The relation between transformational leadership and follower emotional strain: The moderating role of professional ambition

Abstract

The present study tests a model of moderated mediation in the relationship between transformational leadership and follower emotional strain. Based on the job demands-resources model, we suggest that transformational leaders may be able to decrease follower emotional strain by providing social support. It is also proposed that a protective effect of social support from a transformational leader will depend on the employees’ level of professional ambition. Mediation by social support may be stronger for ambitious employees, such that transformational leadership may be associated with less emotional strain for these employees (moderated mediation). A sample of 199 employees participated in a cross-sectional study in Germany. Results confirmed the hypothesized moderated mediation indicating a health-promoting effect of supervisory social support for ambitious employees (not found for low levels of ambition). The study suggests that the idea of a general positive effect of transformational leadership on followers’ emotional strain is not appropriate.

Keywords

Transformational leadership job demands-resources model follower emotional strain follower well-being supervisory social support professional ambition moderated mediation

Prior research on the relation between transformational leadership and followers’ well-being has mainly tended to concentrate on the question of whether transformational leaders promote followers’ well-being and on how transformational leaders achieve this, for example by providing social support (Nielsen and Daniels, 2012; Sosik and Godshalk, 2000) and enhancing work characteristics (Arnold et al., 2007; Nielsen et al., 2008). Less attention has been paid to moderating variables which determine when transformational leadership and followers’ well-being are more closely related (Franke and Felfe, 2011). The present study combines both perspectives in a moderated mediation model, suggesting supervisory social support as a mediator of the relationship between transformational leadership and follower emotional strain, but also investigating whether the strength of this mediation differs for employees with either high or low professional ambition.

Zhu et al. (2009) highlight the need to take into account followers’ characteristics in order to understand differential effects of leadership. Franke and Felfe (2011) demonstrated that organizational commitment is a moderator of the relationship between transformational leadership and follower strain, thereby providing original evidence of the importance of follower boundary conditions for this relationship. The present study focuses on the followers’ level of professional ambition, which is conceptualized as followers’ ‘persistent and generalized striving for success, attainment, and accomplishment’ (Judge and Kammeyer-Mueller, 2012: 759). In their recent article, Judge and Kammeyer-Mueller (2012) demonstrated that ambition is connected to life satisfaction through major life attainments. Based on these findings, they emphasize the utility of studying ambition as a construct within organizational behaviour research. The present study investigates follower ambition in the context of transformational leadership. We focus on social support as a connecting mechanism between transformational leadership and follower well-being, since social support has been proposed to relate to follower motivation (Bakker et al., 2008). Ambitious followers may appreciate their supervisor’s social support to a greater extent than those with less ambition because social support signals the supervisor’s appraisal (Frese, 1999), which is important for promotion (De Andrés et al., 2010), and therefore may be especially meaningful for ambitious followers. Thus, we hypothesize that social support by a transformational leader will result in lower strain levels for followers with high levels of professional ambition. Conversely, for followers with low levels of ambition, a supervisor’s support may not be as important.

As its theoretical basis, our study integrates transformational leadership (Burns, 1978) and the job demands-resources (JD-R) model (Bakker and Demerouti, 2007; Demerouti et al., 2001), suggesting that leaders can influence their subordinates’ emotional strain both by imposing job demands and providing job resources (Tims et al., 2011).

Transformational leadership and follower emotional strain

Transformational leaders guide their followers by giving them an idea of a higher purpose in their work, thereby stimulating their commitment to team goals (Bass and Riggio, 2006). Transformational leadership consists of the four sub-dimensions intellectual stimulation, idealized influence, inspirational motivation, and individual consideration. Most researchers have suggested that transformational leadership can promote follower well-being (see Skakon et al., 2010, for a review). A health-promoting effect of transformational leadership has not only been found in cross-sectional studies (Arnold et al., 2007) but also longitudinally (Nielsen et al., 2008). Even in an experimental setting transformational leadership resulted in lower ratings of followers’ negative affect than transactional leadership (Lyons and Schneider, 2009). By contrast, transformational leadership has also been suggested to potentially impair employees’ well-being (Franke and Felfe, 2011). Rowold and Heinitz (2008) found that transformational leadership resulted in elevated stress levels in a prospective study with a time lag of 6 months. Thus, transformational leadership may have negative consequences in the long run. Other studies have demonstrated that the transformational leadership component intellectual stimulation may relate to elevated burnout levels (Seltzer et al., 1989), while a different study by Stordeur et al. (2001) found that the total transformational leadership score was unrelated to burnout. Interestingly, these studies, which did not confirm a positive relation between transformational leadership and well-being, used negative indicators of follower well-being as opposed to positive indicators, as in other studies (Arnold et al., 2007; Nielsen et al., 2008). Since these results are not without contradiction, and the relation between transformational leadership and health outcomes varies across studies, a closer look at potential moderating variables seems advisable. Skakon and colleagues (2010) have encouraged investigating individual, situational, and relational factors which may account for the observed variance in correlations between leadership variables and followers’ well-being in different studies.

The JD-R model

Examining the relation between leadership and follower well-being, we build on the JD-R model (Bakker and Demerouti, 2007; Demerouti et al., 2001), an extension of the job-demand-control model (Karasek, 1979). According to the JD-R model job-strain results from a workplace situation characterized by high job demands and low job resources. The JD-R model conceptualizes job demands as ‘physical, psychological, social, and organizational aspects of the job that require sustained […] effort or skills and are therefore associated with certain physiological and/or psychological costs’ (Bakker and Demerouti, 2007: 312). Job resources refer to aspects of a job that reduce job demands and the associated physiological and psychological costs, facilitate goal achievement, and/or stimulate personal growth, learning, and development (Schaufeli and Bakker, 2004: 296). Job resources may be located at different organizational levels and include, for example, opportunities for development, feedback, support of one’s supervisor, role clarity, and task characteristics (Bakker and Demerouti, 2007).

The main propositions of the JD-R model suggest that job demands will result in elevated levels of strain and that job resources have a motivating effect (Bakker and Demerouti, 2007). Moreover, job resources have been proposed to buffer the relationship between job demands and job strain (Demerouti et al., 2001). Besides these well-known propositions of the JD-R model, Schaufeli and Bakker (2004) have suggested that poor job resources may also result in elevated strain levels. This effect of job resources on strain has received considerable empirical support (Schaufeli and Bakker, 2004; Schaufeli et al., 2009; Ten Brummelhuis et al., 2011). The present study investigates this link suggesting that social support may be related to lower levels of job strain.

The JD-R model in a leadership context

Leadership may be an important factor for followers’ well-being (Skakon et al., 2010) because leaders have significant influence on both job demands and job resources (Tims et al., 2011), both of which the JD-R model suggests will impact employees’ psychological health (Bakker and Demerouti, 2007). Transformational leadership behaviour explicitly includes the provision of resources (Bass and Riggio, 2006; Tims et al., 2011). Leaders may for example listen to their employees’ problems showing that they care for each of their subordinates (individual consideration). On the other hand, transformational leaders can also impose demands, in the form of high performance expectations towards their employees, for example (inspirational motivation).

Empirically, transformational leadership has been related to various job resources (Tims et al., 2011), which again may relate to reduced strain levels of followers (Schaufeli and Bakker, 2004). For example, positive relations have been found with social support (Nielsen and Daniels, 2012; Sosik and Godshalk, 2000), empowerment (Avolio et al., 2004), meaningful tasks (Arnold et al., 2007; Korek et al., 2010; Nielsen et al., 2008), a high level of job characteristics (Piccolo and Colquitt, 2006; Purvanova et al., 2006), role clarity, and opportunities for development (Nielsen et al., 2008).

The present study examines the effect of transformational leadership on follower emotional irritation (Mohr et al., 2006), a work-related indicator of emotional strain. Mohr and colleagues (2006) define irritation as a state of mental exhaustion. Previous research has found irritation to mediate the relation between stressors and impaired sleep (Berset et al., 2011), psychosomatic symptoms (Höge, 2009), and depression (Dormann and Zapf, 2002). These effects were found in both cross-sectional (Berset et al., 2011) and longitudinal studies (Dormann and Zapf, 2002). Irritation is a sensitive indicator of emotional strain, and therefore best suited to capturing slight deviances from normal well-being that may be more common in the average working population.

As Skakon et al. (2010) conclude, most studies suggest that transformational leadership relates to better well-being of followers (Arnold et al., 2007; Nielsen et al., 2008) because transformational leaders provide a high level of job resources such as good working conditions. Fewer studies have linked transformational leadership to higher stress levels (Rowold and Heinitz, 2008; Seltzer et al., 1989). Therefore, transformational leadership is hypothesized to relate negatively to follower emotional irritation.

H1: Transformational leadership relates negatively to follower emotional irritation.

Supervisory social support as a mediator

The present study investigates supervisory social support as a connecting mechanism between transformational leadership and follower well-being. Social support from the supervisor involves directly helping subordinates, providing affective support and confirming the appropriateness of the subordinate’s actions (Frese, 1999). A negative relationship between social support and psychological health outcomes has been suggested theoretically (Schaufeli and Bakker, 2004) and could be empirically supported in various studies (for a meta-analysis see, Viswesvaran et al., 1999). Two different mechanisms have been suggested which may account for this link (Cohen and Wills, 1985). First, social support may have a direct effect on employee well-being (Van der Doef and Maes, 1999) which may be based on regular positive experiences in a social group. As these kinds of interactions provide reliable social relationships, positive affect and confirm employees’ self-worth, increased well-being is a likely consequence (Cohen and Wills, 1985). Second, social support may also function as a stress buffer alleviating the relation between stressors and strain (Cohen and Wills, 1985; Dormann and Zapf, 1999). This buffer hypothesis suggests that social support may decrease negative effects on individuals’ health, which may be triggered by stressors (Kahn and Byosiere, 1992).

It is important to note that these two mechanisms are not mutually exclusive (Häusser et al., 2010). Some studies supported the proposed mechanisms (Janssen and Nijhuis, 2004; Frese, 1999) but others did not (McClenahan et al., 2007). This variability indicates that third variables may moderate the relationship between social support by the supervisor and follower well-being (Häusser et al., 2010). The issue of moderators will be discussed in the next section.

We propose supervisory social support as a mediator of the relation between transformational leadership and followers’ emotional strain. Although social support has mainly been regarded as a moderator of the relation between job demands and follower strain (Bakker and Demerouti, 2007), our positioning of this variable is in line with previous research demonstrating direct effects of social support on follower well-being (Häusser et al., 2010). As defined by Bass and Riggio (2006), transformational leadership behaviour explicitly includes the provision of social support as it is an aspect of individual consideration. Empirically, transformational leadership was found to relate to higher perceived supervisory social support in an experimental study (Lyons and Schneider, 2009). Supervisory social support, in turn, has been found to lead to lower stress (Sosik and Godshalk, 2000) and a lower likelihood of burnout (Lee and Ashforth, 1996) among followers. The health-promoting effect of supervisory social support has also been found in a longitudinal study (Moyle, 1998) and for objective indicators of psychological dysfunction (Frese, 1999). Moreover, transformational leadership was found to relate to reduced levels of stress (Sosik and Godshalk, 2000) and burnout (Nielsen and Daniels, 2012) through social support. Based on the concept of transformational leadership as defined by Bass and Riggio (2006) and on previous research, transformational leadership is expected to impact follower well-being through social support. Consequently, we hypothesize:

H2: Social support by the supervisor mediates the relation between transformational leadership and irritation.

The moderating role of professional ambition

The present study hypothesizes that transformational leadership may be more closely related to followers’ emotional strain through supervisory social support for employees with high levels of professional ambition. Previous research has demonstrated that subordinates differ with respect to their susceptibility to transformational leadership (Dvir and Shamir, 2003) as well as in regards to their reactions to social support (Beehr et al., 2010). Regarding transformational leadership, Zhu et al. (2009) found a stronger relationship with work engagement for employees who characterized themselves as innovative and willing to take risks. By contrast, Hetland and colleagues suggest that some followers may ‘experience increased stress due to the intense and emotionally charged interactions between leader and subordinates’ (Hetland et al., 2007: 68). Dvir and Shamir (2003) proposed that different receptiveness to transformational leaders may depend on followers’ motivation. For example, not everyone may be willing to go beyond the call of duty as demanded by transformational leaders. Therefore, follower motivation, more precisely professional ambition, is proposed to determine the strength of the mediation of supervisory social support in the relation between transformational leadership and followers’ emotional strain as the supervisor’s support may be of higher relevance for highly motivated followers.

Judge and Kammeyer-Mueller (2012) conceptualize ambition as a middle-level trait (Cantor, 1990), which tend not to be as stable as more distal traits, but also not as situational as goals, behavioural intentions or attitudes. Professional ambition is characterized by a strong motivational component which is also inherent in transformational leadership. Ambitious employees strive to attain work-related goals, and seek professional advancement and promotion (Kieschke and Schaarschmidt, 2008; Schaarschmidt and Fischer, 2003; Voltmer et al., 2011). Schaarschmidt and Fischer (2003) describe professional ambition as an indicator of psychological health, thus differentiating ambition from type A personality (Friedman and Rosenman, 1974). Professional ambition is related to organizational commitment (Allen and Meyer, 1990) and work engagement (Schaufeli et al., 2002) but can be differentiated conceptually. Unlike work engagement, professional ambition does not refer to actual energy and functioning in the workplace (Bakker et al., 2008). It also has a different focus since it is not directed at a specific organization in the way that organizational commitment is (Allen and Meyer, 1990) but rather, captures the individuals’ dedication to their career (Schaarschmidt and Fischer, 2003).

We propose that the extent to which the relationship between transformational leadership and follower emotional irritation is mediated by supervisory social support, may differ for followers with high versus low professional ambition (Figure 1). The moderator professional ambition can either affect path a (the relation between transformational leadership and social support) or path b (the relation between social support and strain), or both. We assume only a moderation of path b. We will first explain why an effect of professional ambition on the relationship between transformational leadership and social support is not expected.

Figure 1.

Moderated mediation: The indirect effect of transformational leadership on emotional irritation through supervisory social support is conditional upon professional ambition.

Transformational leadership is often addressed to the whole group (Korek et al., 2010; Nielsen and Daniels, 2012) by setting group goals, developing a specific group climate or fostering team spirit. It is one of the central tenets of transformational leadership to establish group cohesion, collective efficacy and identity (Shamir et al., 1993). Feinberg et al. (2005) even noted that building consensus is one of the main tasks of transformational leadership. This implies that transformational leaders will treat followers equally and not substantially differ in the amount of social support they give. Therefore, we do not expect path a to be dependent on the level of professional ambition of followers.

By contrast, the role of social support as a mediator between transformational leadership and emotional irritation is expected to differ for followers with high vs. low ambition. This refers to a moderation of path b. Followers with low and high levels of ambition may evaluate the supervisor’s social support differently. That is, individual appraisal processes may account for different relationships between supervisor support and follower strain. The degree to which the supervisor’s behaviour is considered valuable by a follower may depend on individual factors like professional ambition.

Previous research suggests that followers may benefit from transformational leadership to different degrees (Franke and Felfe, 2011; Zhu et al., 2009). We expect the health-promoting effects of transformational leadership to be most apparent for ambitious followers because the leader’s support may be considered more important to them. First, social support by the supervisor indicates that the subordinate is a valued member of the team (Sarason et al., 1996). This appreciation by the supervisor may be especially valued by ambitious followers. Second, as careers are very important to ambitious followers, and the leader has considerable influence on their professional advancement, the leader’s approval may also be more relevant for ambitious followers because it is a precondition of promotion (De Andrés et al., 2010). Moreover, ambitious employees may be especially receptive to transformational leaders because of high goal congruence with their leader (Bono and Judge, 2003). For example, followers reporting high performance orientation were found to evaluate transformational leadership more positively (Ehrhart and Klein, 2001) which might be associated with improved well-being. Being very performance-oriented, this should be true for ambitious followers as well. High importance attached to the leader’s social support in combination with high goal congruence and similarity between leader and follower is hypothesized to result in reduced strain levels for ambitious followers.

By contrast, for followers with low professional ambition the leader’s support may be less important and may thus have a smaller impact on the well-being of these followers. As less ambitious followers are not striving as much for professional advancement (Schaarschmidt and Fischer, 2003), they are less dependent on the leader’s goodwill and positive evaluation which is indicated by supervisory social support (Sarason et al., 1996). Moreover, social support may have detrimental effects under some conditions, for example, when the provision of help represents stressful working conditions (Beehr et al., 2010). We suggest that social support from a transformational leader may at least have less positive effects on less ambitious followers as it is associated with high performance expectations (Bass and Riggio, 2006), which may seem especially stressful to these followers. Furthermore, less ambitious followers may perceive the leader’s support as a means of pushing them to achieve improved performance, which is not their own goal. As a consequence, these followers may not regard the leader’s support as a resource. Thus, the leader’s support may affect follower well-being to a lesser extent.

Concluding, we hypothesize that the followers’ evaluation of the leader’s behaviour (but not the leader’s behaviour itself) may differ according to the followers’ level of professional ambition. Consequently, individual appraisal processes (path b, see Figure 1) but not differential leader behaviour (path a) are proposed to determine the degree to which the follower regards the leader’s support as a resource. We propose a moderated mediation model (Figure 1):

H3: The mediation effect of social support in the relation between transformational leadership and irritation is contingent upon professional ambition. In the case of high ambition, we expect a stronger mediation effect of social support than in case of low ambition.

Method

Participants and procedure

A paper–pencil questionnaire was distributed to employees from financial and service sectors, and from real estate firms in a cross-sectional study in Germany. From a theoretical perspective, a study of professional ambition demands a careful sampling strategy to include both ambitious and less ambitious participants. Organizations practicing an up-or-out strategy, for example, would be inappropriate, as selection processes may be in effect in such organizations that could bias the study’s results. We therefore approached organizations which offer career opportunities but do not claim that their employees strive for professional advancement. In this regard, German public banks were deemed to be an ideal sample. Second, as leadership is the focus of the present study, frequent interaction between team and leader is important. Interactions on a regular basis may be necessary to develop the hypothesized effects of leadership characteristics on followers’ emotional strain. The health-promoting effect of supervisory social support especially demands close and frequent contact. We focused on financial institutions, the service sector, and real estate firms because employees in these fields typically have daily interactions with their leaders. Sixteen of the participating organizations were from the financial sector while three organizations were real estate firms and one organization was from the service sector. The participants were asked to fill in the questionnaire during their working hours. Participation was voluntary and anonymity assured. Small groups of up to 10 employees filled in the questionnaire at the same time in the presence of a researcher. Thus, all participants handed in a questionnaire. Two hundred and eleven questionnaires were returned, of which 199 questionnaires contained no missing values. Participants reported a mean age of 35.87 (SD = 9.96) years and a mean tenure of 116.11 (SD = 73.69) months. Sixty-one percent were female and 39% were male. Thirty-one percent had a university degree.

Measures

Transformational leadership

The present study used the German version (Felfe, 2006) of the Multifactor Leadership Questionnaire (Bass and Avolio, 1995) to measure transformational leadership. This scale consists of 20 items and assesses the four sub-dimensions of transformational leadership: Inspirational motivation, idealized influence (attributed and behaviour), intellectual stimulation, and individual consideration. Previous research has demonstrated these sub-dimensions to be highly interrelated (Judge and Piccolo, 2004). Consequently, they cannot be regarded as independent factors (Carless, 1998). In line with other research (Liaw et al., 2010), the present study therefore utilises the total score for the construct. The reliability figures indicate very good internal consistency (Cronbach’s α = 0.93). Responses were rated on a 5-point Likert scale ranging from 1 (never) to 5 (regularly, almost always).

Irritation

Emotional strain was measured by the emotional irritation scale (Mohr et al., 2005), which consists of five items. A sample item is ‘I get irritated easily, although I don’t want this to happen’. Regarding construct validity, irritation has been found to relate positively to stressors (Mohr et al., 2006). Moreover, irritation has also been demonstrated as an antecedent of depression (Dormann and Zapf, 2002), psychosomatic symptoms (Höge, 2009), and impaired sleep (Berset et al., 2011). The irritation scale displayed good reliability in the present sample (Cronbach’s α = 0.81). It was combined with a 7-point Likert scale ranging from 1 (not true at all) to 7 (almost completely true).

Social support by supervisor

Participants indicated the degree of work-related social support they receive from their supervisor. A scale by Frese (1989) was used consisting of five items. A sample item is ‘How much is your supervisor willing to listen to your work-related problems?’ The scale displayed good reliability in the present study (Cronbach’s α = 0.83). Participants rated the social support provided by their supervisor on a 4-point Likert scale ranging from 1 (not at all) to 4 (completely).

Professional ambition

The measure of professional ambition employed in the present study was taken from the AVEM (Schaarschmidt and Fischer, 2003), an instrument for the assessment of health-relevant behaviour and experiences in the workplace. High levels of ambition are related to psychological health and distinguished from the risk pattern known as type A personality (Friedman and Rosenman, 1974) which is associated with a lack of detachment from work and a susceptibility to risking depletion of resources. We used the four items with the highest item-total-correlations as reported by Schaarschmidt and Fischer (2003). A sample item is: ‘I want to achieve more in my career than most people I know’. Internal consistency indicated good reliability (Cronbach’s α = 0.87). Likert response categories ranged from 1 (not at all true) to 5 (completely true).

Control variables

In order to control for possible biases, demographic data were also collected which may impact followers’ emotional strain; the study controlled for the effects of sex, age, weekly working hours, and education. Sex was coded 0 for women and 1 for men. As men indicated significantly higher professional ambition than women in our sample, we included sex as a control variable in our analyses. Controlling for the effects of age was deemed appropriate in the context of health. We also controlled for weekly working hours as a proxy for work demands, which is mostly unable to be influenced by the leader. Finally, the study controlled for the effect of education since education may relate to professional ambition and followers’ emotional strain. Participants indicated their highest educational level. According to the tripartite German educational system, education was classified as follows: 1 = basic secondary school, 2 = secondary school, 3 = higher secondary school, 4 = vocational education, 5 = vocational academy, 6 = university.

Analyses

As transformational leadership and supervisory social support are conceptually related, a confirmatory factor analysis was computed using the software Amos 20.0 (Arbuckle, 2011) to ensure that they can be treated as two distinct concepts. We compared a one-factor model (χ² (275) = 770.73, CFI = 0.79, RMSEA = .10) to a two-factor model which treated transformational leadership and social support by the supervisor as distinct but related constructs (χ² (274) = 653.64, CFI = .84, RMSEA = 0.08). A chi-square difference test indicated that the two-factor model achieved a significantly better fit to the data than the more parsimonious one-factor model (Δχ² (1) = 117.09, p < 0.001). Although the fit of the two-factor-model is not good, the results indicate that transformational leadership and supervisory support are distinct concepts. Moreover, as the most critical point is to demonstrate that individual consideration can be differentiated from supervisory social support, we additionally compared a one-factor model (χ² (27) = 86.01, CFI = .91, RMSEA = .11) to a two-factor model (χ² (26) = 51.26, CFI = .96, RMSEA = 0.07) in which items referring to individual consideration and items referring to supervisory support loaded on separate factors. The two-factor model displayed significantly better fit (Δχ² (1) = 34.75, p < 0.001). Consequently, it can be concluded that both transformational leadership, as a compound measure, as well as individual consideration can be distinguished from supervisory support.

Statistical approach

All hypotheses were tested using regression analysis in combination with bootstrapping, a statistical re-sampling method. Bootstrapping involves the drawing of a large number of subsamples from the original sample, within all of which the analyses are performed. The number of re-samples was set to N = 5000 as suggested by Hayes (2009). Hypothesis 2 was tested by applying a macro for indirect effects by Preacher and Hayes (2008). Contrary to the classic causal step procedure (Baron and Kenny, 1986), the procedure by Preacher and Hayes (2008) can directly estimate indirect effects. Consequently, an indirect effect can be detected even when a direct effect is missing (Hayes, 2009). Confidence intervals for the true score of the indirect effect are specified and the null hypothesis is rejected if the confidence interval excludes zero. Since bootstrapping does not assume a normal distribution of the data, it has been demonstrated to result in more accurate confidence intervals of indirect effects compared to the classic stepwise procedure (MacKinnon et al., 2002). Hypothesis 3 proposed that the effect of transformational leadership on irritation would be mediated by social support and that this mediation effect would be moderated by professional ambition (moderated mediation, see Figure 1). Again a macro for SPSS (Preacher et al., 2007) was applied to estimate the moderated mediation involving a bootstrap procedure. In a moderated mediation either path a or path b or both may be moderated by a third variable. Significance of the conditional indirect effect is indicated by the bootstrap confidence interval.

Results

Means, standard deviations, correlations, and internal consistencies for all study variables are presented in Table 1. All regression analyses controlled for age, gender, education, and weekly working hours as these variables may relate to occupational health. Moreover, all predictors were mean centred as suggested by Aiken and West (1991) to avoid multicollinearity with their product terms.

Table 1.

Means, standard deviations, correlations, and internal consistencies for all study variables.

Variable	Mean	SD	1	2	3	4	5	6	7
1	Transformational leadership	3.64	0.58	(.93)
2	Professional ambition	3.28	0.87	.33	(.87)
3	Supervisor social support	3.24	0.53	.64	.07	(.83)
4	Emotional irritation	2.15	0.87	−.08	−.06	−.15	(.81)
5	Age	35.87	9.96	.09	−.22	.21	−.11
6	Gender^a	0.37	0.48	.14	.25	.09	−.05	−.05
7	Working hours per week	38.27	3.30	.23	.27	.17	−.05	−.02	.25
8	Education	3.65	1.38	−.02	.17	−.03	−.19	.00	.20	−.20

Note. N = 199. Numbers in parenthesis are reliabilities where appropriate. ^a0 = female; 1 = male. For |r| ≥ .15 p < .05, for |r| ≥ .19 p < .01, for |r| ≥ .25 p < .001 (two-sided).

In order to test the first hypothesis a regression analysis was computed. Irritation was regressed on transformational leadership. Results are presented in Table 2. Contrary to expectations, the constructs were unrelated and the confidence interval included zero. Thus hypothesis 1 was not supported.

Table 2.

Regression results for mediation model.

Variable	Irritation
Variable	B	95% CI [LL; UL]
Transformational leadership on supervisory social support (path a)	.57***	[.477; .660]
Supervisory social support on irritation (path b)	−.25^†	[−.473; .020]
Transformational leadership on irritation (path c – total effect)	−.12	[−.326; .109]
Transformational leadership on irritation (path c’ – direct effect)	.03	[−.211; .279]
R²	.08
Adjusted R²	.05
Bootstrapping results for indirect effect	B	95% BC CI [LL; UL]
Transformational leadership – social support - irritation	−.145	[−.300; .013]

Note. N = 199. BC = Bias corrected. CI = Confidence intervals. LL = Lower level, UL = Upper Level.

Control variables in the model: Age, gender, working hours per week, education. ^†p ≤ .1; ***p ≤ .001 (two-sided).

Second, it was proposed that supervisory social support would mediate the relation between transformational leadership and irritation. The hypothesis was assessed using a macro for indirect effects (Preacher and Hayes, 2008). As the confidence interval contained zero, mediation by supervisory social support was not supported. Though path a of the mediation model was significant (B = 0.57, p < 0.001), path b was not (B = -0.25, n.s.). As previously mentioned, a direct effect of transformational leadership on irritation (path c) was not found. Nevertheless, according to Preacher and Hayes (2008) a significant path c is not necessary to support mediation. Results are presented in Table 2.

Thirdly, we hypothesized that the mediation by supervisory social support would be conditional upon professional ambition. Thus, for ambitious employees social support by a transformational leader is hypothesized to impact irritation to a greater degree. To test this hypothesis, a moderated mediation analysis was computed to estimate the conditional indirect effect of transformational leadership on irritation (Preacher et al., 2007). The conditional indirect effect determines whether different levels of professional ambition influence the strength of the mediation. High and low levels of professional ambition were operationalized as one standard deviation below and above the mean score as suggested by Preacher et al. (2007). First, we assessed path a of the mediation model (see Figure 1) as being conditional upon professional ambition to examine if – against our assumption – less ambitious employees feel less supported by their leader. We estimated the indirect effect using a bootstrap procedure. As expected, the indirect effect did not vary for different levels of professional ambition indicating that path a of the mediation model (see Figure 1) was not moderated by professional ambition. Thus, the procedure for a moderated mediation of path b (effect of supervisory social support on irritation) was applied. Results indicate a significant conditional indirect effect for high professional ambition (B = -0.27, p < 0.05). The bootstrap 95% confidence interval excluded zero when professional ambition was high ( + 1 SD) and when it was medium but not in the case of low ambition (- 1 SD). Thus the mediating effect of supervisory social support between transformational leadership and irritation applies only to employees with medium to high professional ambition (see Figure 2). Consequently, hypothesis 3 is supported. Estimates and significance levels are displayed in Table 3.

Figure 2.

Interaction between supervisory social support and professional ambition.

Table 3.

Regression results for conditional indirect model.

Model	Irritation B
Transformational leadership (TL)	.06
Professional ambition	−.06
Supervisor social support	−.27^†
Support x ambition		−.21^†
Irritation
Bootstrapping results for conditional indirect effect	B	BC 95% CI [LL; UL]
Professional ambition
−1 SD (2.41)	−.038	[−.258; .176]
M (3.28)	−.155	[−.318; −.005]
+1 SD (4.14)	−.273	[−.533; −.041]

Note. N = 199. BC = Bias corrected. CI = Confidence intervals. LL = Lower level, UL = Upper Level.

Control variables in the model: Age, gender, working hours per week, education.

†

p ≤ .1 (two-sided).

Discussion

The JD-R model suggests that leaders can impact their subordinates’ emotional strain by imposing job demands and providing job resources (Schaufeli and Bakker, 2004; Schaufeli et al., 2009; Ten Brummelhuis et al., 2011). Though both negative (Tepper, 2000; Pelletier, 2011) and positive effects (Nielsen et al., 2008) of a leader’s behaviour have been reported in the literature, transformational leadership has so far mainly been linked to positive effects on followers’ well-being (Skakon et al., 2010). Results from this study demonstrate a more detailed picture and indicate that transformational leadership is not necessarily positively related to follower well-being for all followers.

In line with previous research (Skakon et al., 2010), it was hypothesized that transformational leadership would be positively linked to follower well-being and thus negatively related to irritation, a work-related indicator of emotional strain (Mohr et al., 2006). After controlling for age, gender, education, and weekly working hours, transformational leadership and irritation were practically unrelated. Consequently, the first hypothesis was rejected. This result differs from findings of previous studies on the relationship between transformational leadership and follower well-being as reviewed by Skakon et al. (2010). It may be due to the operationalization of irritation as a negative indicator of follower well-being as opposed to the positive operationalizations in other studies (e.g. Arnold et al., 2007; Nielsen et al., 2008). Stordeur and colleagues (2001) also found, when they utilised a negative indicator of well-being, that transformational leadership was unrelated to organizational stress. Moreover, this result is not at odds with the implications of the JD-R model. It may depend on subjective evaluation as to whether subordinates mainly focus on the demands or the resources provided by a transformational leader. Thus, some variability in relationships can be expected.

Second, the present study examined supervisory social support as a mediator between transformational leadership and follower irritation. Transformational leadership explicitly involves the provision of social support (Bass and Riggio, 2006). In line with the JD-R model (Demerouti et al., 2001; Schaufeli and Bakker, 2004) and previous research (Nielsen and Daniels, 2012; Sosik and Godshalk, 2000), the present study hypothesized that transformational leaders may influence their followers’ emotional strain by providing social support. Results reveal that although transformational leadership was significantly related to supervisory social support, supervisory social support was unrelated to followers’ irritation. Thus, the hypothesized mediation of supervisory social support was not confirmed. Previous research on the relation between social support and follower well-being has revealed mixed results. A meta-analysis by Viswesvaran et al. (1999) only demonstrated a weak relation. Moreover, although a number of studies have found support for a main effect of social support on follower well-being, this effect has not been found in other studies (Van der Doef and Maes, 1999; Häusser et al., 2010). The ambiguity of empirical findings regarding the relationship between transformational leadership and follower well-being, as well as the hypothesized mediation by supervisory social support, call for research on moderators. Dvir and Shamir (2003) have suggested followers’ different receptiveness to transformational leaders may depend on motivational processes. Consequently, the present study investigated a moderator with a strong motivational component, professional ambition.

The third hypothesis proposed that the mediation by supervisory social support may depend on the extent of professional ambition. Two different processes may account for this moderated mediation: First, highly ambitious followers might perceive transformational leadership to be more supportive than less ambitious followers (path a in Figure 1). Second, the leader’s support may be differently evaluated by the followers, which means that individual appraisal processes may account for this outcome (path b in Figure 1).

We hypothesized that the latter process would be relevant in the context of professional ambition as transformational leaders are defined as supporting all followers (Bass and Riggio, 2006). Path a of the mediation model (Figure 1) was not found to be conditional upon professional ambition. Thus evidence of the first process was not found. However, a moderated mediation analysis of path b (Preacher et al., 2007) found support for the second process. The relation between transformational leadership and follower irritation was mediated by supervisory social support for medium and highly ambitious employees. However, this was not true for followers with low professional ambition (see Figure 2), thereby confirming hypothesis 3. The same pattern of results was revealed for all four facets of transformational leadership.

The moderation of path b of the mediation model indicates that individual evaluation or appraisal processes seemingly account for the effect leading to less irritation among highly ambitious employees. Professional ambition may determine the degree to which the supervisor’s supporting behaviour is perceived as a resource by the follower. Ambitious followers might perceive supervisory support as a helpful acknowledgement which emphasizes that the leader appreciates the follower. This positive effect may again result in lower levels of strain. Though the supervisor’s support was confirmed as a mediator for ambitious followers, it was not a full mediation indicating that other mediating variables may be existent.

Conversely, less ambitious subordinates may experience the supervisor’s support as less helpful, so that the leader’s behaviour may not be an equally important resource for them. Followers with low professional ambition may even feel pressured by their supervisor to produce superior performance. Moreover, it has been demonstrated that social support may have adverse effects, for example when help represents demanding working conditions (Beehr et al., 2010). Although no negative effects of supervisory social support were found in this study, the results indicate that social support is not always connected to increased well-being.

When comparing our results to the study by Franke and Felfe (2011), the results referring to the moderator apparently differ. While the present study found high professional ambition to amplify the mediation effect, Franke and Felfe (2011) found the opposite pattern when analyzing organizational commitment as a moderator between transformational leadership and strain. We argue that the different focus of the moderators (ambition focuses on career goals while organizational commitment focuses on the current employer) can account for this divergent finding: Highly ambitious employees can of course be almost uncommitted to their organization if they, for instance, perceive that their current employer does not offer opportunities to fulfil their professional goals. Differently, employees with low ambition can display strong organizational commitment when their career goals are fulfilled by their employer. Taken together, both constructs denote different work-related aspects and their effect in empirical studies can thoroughly differ.

Dvir and Shamir (2003) proposed that employees may be differentially receptive to transformational leadership. Zhu et al. (2009) found that innovative followers who were willing to take risks profited more from transformational leadership than other followers. The present study suggests that highly ambitious followers may also be especially well suited to transformational leadership, as they seem to appreciate leader’s social support more than less ambitious followers. It has previously been demonstrated that followers with high levels of extraversion, an internal locus of control, and low neuroticism report higher levels of social support (Chay, 1993; Swickert et al., 2010). Nevertheless, the present study did not reveal any differences with respect to the perceived level of supervisory social support for followers with high and low professional ambition. This is in line with theory suggesting that transformational leaders will value and support each of their followers (Bass and Riggio, 2006). Consequently, a different tendency to get support from the supervisor does not account for our results.

Moreover, the support for the moderation by professional ambition may partly explain the rejection of hypothesis 1 and 2, as a positive relation between transformational leadership and emotional strain as well as the proposed mediation by supervisory social support only holds true for a subsample but is different for other employees. It can be concluded that individual differences between subordinates, such as for example different levels of ambition, should be considered when applying the JD-R model in a leadership context.

Limitations and strengths

A number of limitations have to be taken into consideration when interpreting the results of this study. First, due to the cross-sectional design, causal inferences cannot be made from the present study. However, well-founded longitudinal research assessing relationships between job characteristics and well-being (De Jonge et al., 2001; Ter Doest and De Jonge, 2006) as well as between leadership variables and well-being (Nielsen et al., 2008) suggest that models of regular causation (i.e. leadership/job characteristics impact well-being) achieve a better fit than models of reverse or reciprocal causation. These studies support our interpretation. Likewise, causal interpretation of the mediation by supervisory social support based on cross-sectional data is not appropriate. Nevertheless, previous longitudinal research has found that follower stress can be reduced by the leader’s social support (Moyle, 1998).

Common method bias may also have affected the results of this study as only one method and one data source were employed. However, according to Chan (2009) self-reports may be the best way to assess private events. Likewise, the individual employees should be best suited to judge their own professional ambition because they know best about their future plans and goals. The same applies to emotional strain. Nevertheless, other measures of strain including physiological indicators could be considered an alternative measure, in future replications of these results. Transformational leadership and social support by the supervisor can be assessed by both leader and subordinate though typically judged by the subordinate. We chose to rely on the subordinates’ perspective of their leaders’ behaviour and social support because the subordinates’ individual perception is important for their emotional strain. Moreover, evaluation by the leader may be subject to demand characteristics which again may bias the study’s results. As recommended by Conway and Lance (2010) we used established scales with acceptable internal consistency for which construct validity has been demonstrated. Furthermore, confirmatory factor analysis indicated that transformational leadership and supervisory social support can be regarded as two distinct factors (see above).

Despite of these limitations, the present study lends initial support to professional ambition as a moderator of the relationship between transformational leadership and follower emotional strain. With respect to the sampling strategy, we focused on employees in banks, insurance, service and real estate firms because followers are likely to interact on an almost daily basis with their leaders in these occupational fields and frequent interaction makes the effects of leadership characteristics on follower emotional strain more likely. Second, we aimed to include organizations which are attractive for both ambitious and less ambitious employees to ensure sufficient variance in professional ambition. We therefore excluded organizations practicing an up-or-out strategy. As can be seen from Table 1, the variance of professional ambition was comparable to other scales not indicating restricted variance.

Implications

Over recent years, research has concentrated on global, positive effects of transformational leadership and on general mediating variables. The results of the present study suggest that there is need for a more individualistic view of the effects of transformational leadership. Characteristics of dyads of leaders and subordinates (Graen and Uhl-Bien, 1995) should be taken into account in considering effects of transformational leadership. Furthermore, as not all employees are equally affected by a transformational leader (Dvir and Shamir, 2003), it may also be worthwhile taking into account the individual characteristics of the subordinate, such as professional ambition. These individual characteristics may determine whether the transformational leader’s behaviour is perceived as a resource by the followers which in turn may account for different reactions to a transformational leader. The present study highlights that transformational leadership may not have positive effects on the emotional strain of all followers. Consequently, it seems important for leaders to consider carefully which leadership behaviour may be appropriate for which of their subordinates. For example, supervisory social support appears not to be an effective way to promote follower well-being for followers with low professional ambition.

In the present study transformational leadership and follower professional ambition were significantly correlated (r = .33, p < 0.01, see Table 1). This may reflect the tendency of ambitious employees to evaluate their leader as more transformational, or the tendency of leaders’ to put more effort into ambitious team members.

On the other hand, transformational leaders may also promote professional ambition. Shamir et al. (1993) have argued that transformational leaders stimulate their followers’ motivation. But while transformational leaders emphasize high moral standards, collectivism and a common purpose, high professional ambition includes a strong focus on individual professional advancement. Nevertheless, future research should try to disentangle the relationship between transformational leadership and professional ambition to find out if these constructs are causally related. Moreover, the stability of professional ambition deserves some attention (Judge and Kammeyer-Mueller, 2012). Both these issues call for longitudinal research. Future research may also focus on goal congruence between leader and subordinate and investigate the fit between follower and leader. The results of this study also indicate that there may be no general health-promoting mode of leadership. Rather, interventions for teaching health-promoting leadership should consider the individual motives of employees. As a consequence, training programs of health-promoting leadership should be tailored to each team and team member.

Footnotes

Funding

This research was supported by a grant from the Deutsche Forschungsgemeinschaft (DFG), grant number: Mo 440/4-1 (2).

Author biographies

Torsten J Holstad is a research associate at the University of Leipzig, Germany. He earned his PhD in the field of work and organizational psychology at the University of Leipzig in 2014. His research focus is on leadership, psychological wellbeing, organizational justice, and occupational health interventions. He also works as a coach, trainer, and consultant.

Sabine Korek is a research associate at the University of Leipzig. She did her PhD in work and organizational psychology at the University of Leipzig in 2013. Her research focus is on leadership, occupational health, and gender.

Thomas Rigotti is a professor of Work, Organizational and Business Psychology at the Johannes Gutenberg-University Mainz, Germany. He received his PhD in 2008 from the University of Leipzig, and became full professor from 2013. His research interests include consequences of atypical employment, health-promoting leadership, organizational justice, work interruptions, as well as stress, and health at work.

Gisela Mohr is a professor of work and organizational psychology at the University of Leipzig. She did her PhD at the University of Osnabrück, Germany, in 1985. Her main research focus is on stress, work and health, leadership, unemployment, job insecurity and gender.

References

Aiken

West

(1991) Multiple regression: Testing and interpreting interactions, Thousand Oaks, Calif: Sage.

Allen

Meyer

(1990) Measurement and antecedents of affective, continuance and normative commitment to the organization. Journal of Occupational Psychology 63: 1–18.

Arbuckle

(2011) IBM SPSS Amos 20 User's Guide, Chicago, IL: Amos Development Corporation.

Arnold

Turner

Barling

Kelloway

McKee

(2007) Transformational leadership and psychological well-being: The mediating role of meaningful work. Journal of Occupational Health Psychology 12: 193–203.

Avolio

Zhu

Koh

Bhatia

(2004) Transformational leadership and organizational commitment: mediating role of psychological empowerment and moderating role of structural distance. Journal of Organizational Behavior 25: 951–968.

Bakker

Demerouti

(2007) The job demands-resources model: State of the art. Journal of Managerial Psychology 22: 309–328.

Bakker

Schaufeli

Leiter

Taris

(2008) Work engagement: An emerging concept in occupational health psychology. Work & Stress 22: 187–200.

Baron

Kenny

(1986) The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology 51: 1173–1182.

Bass

Avolio

(1995) MLQ Multifactor Leadership Questionnaire. Technical Report, Redwood City, CA: Mind Garden.

10.

Bass

Riggio

(2006) Transformational leadership, New Jersey: Lawrence Erlbaum Associates.

11.

Beehr

Bowling

Bennett

(2010) Occupational Stress and Failures of Social Support: When Helping Hurts. Journal of Occupational Health Psychology 15: 45–59.

12.

Berset

Elfering

Luehty

Luethi

Semmer

(2011) Work Stressors and Impaired Sleep: Rumination as a Mediator. Stress and Health 27: E71–E82.

13.

Bono

Judge

(2003) Self-concordance at work: Toward understanding the motivational effects of transformational leaders. Academy of Management Journal 46: 554–571.

14.

Burns

(1978) Leadership, New York: Harper Row.

15.

Cantor

(1990) From thought to behavior: “Having” and “doing” in the study of personality and cognition. American Psychologist 45: 735–750.

16.

Carless

(1998) Assessing the discriminant validity of transformational leader behaviour as measured by the MLQ. Journal of Occupational and Organizational Psychology 71: 353–358.

17.

Chan

(2009) So why ask me? Are self-report data really that bad? In: Lance

Vandenberg

(eds) Statistical and methodological myths and urban legends: Doctrine, verity and fable in the organizational and social sciences, New York: Routledge, pp. 311–338.

18.

Chay

(1993) Social support, individual differences and well-being: A study of small business entrepreneurs and employees. Journal of Occupational and Organizational Psychology 66: 285–302.

19.

Cohen

Wills

(1985) Stress, social support, and the buffering hypothesis. Psychological Bulletin 98: 310–357.

20.

Conway

Lance

(2010) What Reviewers Should Expect from Authors Regarding Common Method Bias in Organizational Research. Journal of Business and Psychology 25: 325–334.

21.

De Andrés

García-Lapresta

González-Pachón

(2010) Performance appraisal based on distance function methods. European Journal of Operational Research 207: 1599–1607.

22.

De Jonge

Dormann

Janssen

PPM

Dollard

Landeweerd

Nijhuis

FJN

(2001) Testing reciprocal relationships between job characteristics and psychological well-being: A cross-lagged structural equation model. Journal of Occupational and Organizational Psychology 74: 29–46.

23.

Demerouti

Bakker

Nachreiner

Schaufeli

(2001) The job demands-resources model of burnout. Journal of Applied Psychology 86: 499–512.

24.

Dormann

Zapf

(1999) Social support, social stressors at work, and depressive symptoms: Testing for main and moderating effects with structural equations in a three-wave longitudinal study. Journal of Applied Psychology 84: 874–884.

25.

Dormann

Zapf

(2002) Social stressors at work, irritation, and depressive symptoms: Accounting for unmeasured third variables in a multi-wave study. Journal of Occupational and Organizational Psychology 75: 33–58.

26.

Dvir

Shamir

(2003) Follower developmental characteristics as predicting transformational leadership: a longitudinal field study. Leadership Quarterly 14: 327–344.

27.

Ehrhart

Klein

(2001) Predicting followers' preferences for charismatic leadership: The influence of follower values and personality. Leadership Quarterly 12: 153–179.

28.

Feinberg

Ostroff

Burke

(2005) The role of within-group agreement in understanding transformational leadership. Journal of Occupational and Organizational Psychology 78: 471–488.

29.

Felfe

(2006) Validierung einer deutschen Version des “Multifacor Leadership Questionnaire” (MLQ Form 5 x Form Short) von Bass und Avolio (1995) [Validation of the German version of the “Multifactor Leadership Questionnaire” (MLQ Form 5 x Short) by Bass and Avolio (1995)]. Zeitschrift für Arbeits- und Organisationspsychologie 50: 61–78.

30.

Franke

Felfe

(2011) How does transformational leadership impact employees' psychological strain? Examining differentiated effects and the moderating role of affective organizational commitment. Leadership 7: 295–316.

31.

Frese

(1989) Gütekriterien der Operationalisierung von sozialer Unterstützung am Arbeitsplatz [Reliability and validity of an operationalization of social support at work]. Zeitschrift für Arbeitswissenschaft 43: 112–121.

32.

Frese

(1999) Social support as a moderator of the relationship between work stressors and psychological dysfunctioning: A longitudinal study with objective measures. Journal of Occupational Health Psychology 4: 179–192.

33.

Friedman

Rosenman

(1974) Type A Behaviour and Your Heart, New York, NY: Knopf.

34.

Graen

Uhl-Bien

(1995) Relationship-based approach to leadership: Development of leader-member exchange (LMX) theory of leadership over 25 years: Applying a multi-level multi-domain perspective. Leadership Quarterly 6: 219–247.

35.

Häusser

Mojzisch

Niesel

Schulz-Hardt

(2010) Ten years on: A review of recent research on the Job Demand-Control (-Support) model and psychological ‘well-being'. Work & Stress 24: 1–35.

36.

Hayes

(2009) Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs 76: 408–420.

37.

Hetland

Sandal

Johnsen

(2007) Burnout in the information technology sector: Does leadership matter? European Journal of Work and Organizational Psychology 16: 58–75.

38.

Höge

(2009) When work strain transcends psychological boundaries: an inquiry into the relationship between time pressure, irritation, work-family conflict and psychosomatic complaints. Stress and Health 25: 41–51.

39.

Janssen

Nijhuis

(2004) Associations Between Positive Changes in Perceived Work Characteristics and Changes in Fatigue. Journal of Occupational and Environmental Medicine 46: 866–875.

40.

Judge

Kammeyer-Mueller

(2012) On the Value of Aiming High: The Causes and Consequences of Ambition. Journal of Applied Psychology 97: 758–775.

41.

Judge

Piccolo

(2004) Transformational and transactional leadership: A meta-analytic test of their relative validity. Journal of Applied Psychology 89: 755–768.

42.

Kahn

Byosiere

(1992) Stress in Organizations. In: Dunnette

Hough

(eds) Handbook of industrial and organizational psychology, Palo Alto: Consulting Psychologists Press, pp. 571–650.

43.

Karasek

(1979) Job demands, job decision latitude, and mental strain: Implications for job redesign. Administrative Science Quarterly 24: 285–308.

44.

Kieschke

Schaarschmidt

(2008) Professional commitment and health among teachers in Germany: A typological approach. Learning and Instruction 18: 429–437.

45.

Korek

Felfe

Zaepernick-Rothe

(2010) Transformational leadership and commitment: A multilevel analysis of group-level influences and mediating processes. European Journal of Work and Organizational Psychology 19: 364–387.

46.

Lee

Ashforth

(1996) A meta-analytic examination of the correlates of the three dimensions of job burnout. Journal of Applied Psychology 81: 123–133.

47.

Liaw

Chi

Chuang

(2010) Examining the Mechanisms Linking Transformational Leadership, Employee Customer Orientation, and Service Performance: The Mediating Roles of Perceived Supervisor and Coworker Support. Journal of Business and Psychology 25: 477–492.

48.

Lyons

Schneider

(2009) The effects of leadership style on stress outcomes. Leadership Quarterly 20: 737–748.

49.

MacKinnon

Lockwood

Hoffman

West

Sheets

(2002) A comparison of methods to test mediation and other intervening variable effects. Psychological Methods 7: 83–104.

50.

McClenahan

Giles

Mallett

(2007) The importance of context specificity in work stress research: A test of the Demand-Control-Support model in academics. Work & Stress 21: 85–95.

51.

Mohr

Müller

Rigotti

(2005) Normwerte der Skala Irritation: Zwei Dimensionen psychischer Beanspruchung [Standardisation data of the Irritation Scale. Two dimensions of mental strain]. Diagnostica 51: 12–20.

52.

Mohr

Müller

Rigotti

Aycan

Tschan

(2006) The assessment of psychological strain in work contexts. European Journal of Psychological Assessment 22: 198–206.

53.

Moyle

(1998) Longitudinal influences of managerial support on employee well-being. Work & Stress 12: 29–49.

54.

Nielsen

Daniels

(2012) Does shared and differentiated transformational leadership predict followers' working conditions and well-being? Leadership Quarterly 23: 383–397.

55.

Nielsen

Randall

Yarker

Brenner

(2008) The effects of transformational leadership on followers' perceived work characteristics and psychological well-being: A longitudinal study. Leadership Quarterly 22: 16–32.

56.

Pelletier

(2011) Leader toxicity: An empirical investigation of toxic behavior and rhetoric. Leadership 6: 373–389.

57.

Piccolo

Colquitt

(2006) Transformational leadership and job behaviors: The mediating role of core job characteristics. Academy of Management Journal 49: 327–340.

58.

Preacher

Hayes

(2008) Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods 40: 879–891.

59.

Preacher

Rucker

Hayes

(2007) Assessing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research 42: 185–227.

60.

Purvanova

Bono

Dzieweczynski (2006) Transformational leadership, job characteristics, and organizational citizenship performance. Human Performance 19: 1–22.

61.

Rowold

Heinitz

(2008) Führungsstile als Stressbarrieren [Leadership styles as stress barriers]. Zeitschrift für Personalpsychologie 7: 129–140.

62.

Sarason

Brock

Pierce

(1996) Social support: Current status, current issues. In: Spielberger

Sarason

(eds) Stress and emotions: Anxiety, anger, and curiosity, Washington, D.C: Taylor & Francis, pp. 3–27.

63.

Schaarschmidt

Fischer

(2003) AVEM - Arbeitsbezogenes Verhaltens- und Erlebensmuster [AVEM - Patterns of work-related coping behavior], Frankfurt, Germany: Swets & Zeitlinger.

64.

Schaufeli

Bakker

(2004) Job demands, Job resources, and their relationship with burnout and engagement: A multi-sample study. Journal of Organizational Behavior 25: 293–315.

65.

Schaufeli

Bakker

van Rhenen

(2009) How changes in job demands and resources predict burnout, work engagement, and sickness absenteeism. Journal of Organizational Behavior 30: 893–917.

66.

Schaufeli

Martinez

Pinto

Salanova

Bakker

(2002) Burnout and Engagement in University Students: A Cross-National Study. Journal of Cross-Cultural Psychology 33: 464–481.

67.

Seltzer

Numerof

Bass

(1989) Transformational Leadership: Is it a source of more burnout and stress? Journal of Health and Human Resources Administration 12: 174–185.

68.

Shamir

House

Arthur

(1993) The motivational effects of charismatic leadership: A self-concept based theory. Organization Science 4: 577–594.

69.

Skakon

Nielsen

Borg

Guzman

(2010) Are leaders' well-being behaviours and style associated with the affective well-being of their employees? A systematic review of three decades of research. Work & Stress 24: 107–139.

70.

Sosik

Godshalk

(2000) Leadership styles, mentoring functions received, and job-related-stress: A conceptual model and preliminary study. Journal of Organizational Behavior 21: 365–390.

71.

Stordeur

D'hoore

Vandenberghe

(2001) Leadership, organizational stress, and emotional exhaustion among hospital nursing staff. Journal of Advanced Nursing 35: 533–542.

72.

Swickert

Hittner

Foster

(2010) Big Five traits interact to predict perceived social support. Personality and Individual Differences 48: 736–741.

73.

Ten Brummelhuis

Ter Hoeven

Bakker

Peper

(2011) Breaking through the loss cycle of burnout: The role of motivation. Journal of Occupational and Organizational Psychology 84: 268–287.

74.

Tepper

(2000) Consequences of abusive supervision. The Academy of Management Journal 43: 178–190.

75.

Ter Doest

De Jonge

(2006) Testing causal models of job characteristics and employee well-being: A replication study using cross-lagged structural equation modeling. Journal of Occupational and Organizational Psychology 79: 499–507.

76.

Tims

Bakker

Xanthopoulou

(2011) Do transformational leaders enhance their followers' daily work engagement? Leadership Quarterly 22: 121–131.

77.

Van der Doef

Maes

(1999) The Job Demand-Control (-Support) Model and psychological well-being: A review of 20 years of empirical research. Work & Stress 13: 87–114.

78.

Viswesvaran

Sanchez

Fisher

(1999) The Role of Social Support in the Process of Work Stress: A Meta-Analysis. Journal of Vocational Behavior 54: 314–334.

79.

Voltmer

Spahn

Schaarschmidt

Kieschke

(2011) Work-related behavior and experience patterns of entrepreneurs compared to teachers and physicians. International Archives of Occupational and Environmental Health 84: 479–490.

80.

Zhu

Avolio

Walumbwa

(2009) Moderating Role of Follower Characteristics with Transformational Leadership and Follower Work Engagement. Group & Organization Management 34: 590–619.