Keywords

1 Introduction

Due to corporate scandals (e.g., Enron, WorldCom or Anglo Irish Bank) and the associated unethical behaviors of leaders and employees, public confidence into leader behavior has declined in the recent years [35]. Indeed, some studies have argued that certain leadership styles may be associated with unethical behaviors [6, 29, 32, 41, 54]. Thus, recent research has focused on leadership styles that encourage socially responsible and moral behaviors such as servant leadership, ethical leadership, and authentic leadership [2, 3, 11].

Among other outcomes, previous research has also identified leadership as a determinant of employees’ psychological health [33]. For example, in their 2011 report, the Federal Institute for Occupational Safety and Health in Germany reported a steady decrease of work-related psychological health in the past years and predicted that in the upcoming years this trend will result in high absenteeism rates in Germany [12]. Thus, from an organizational perspective, threats to employees’ psychological health can be expected to account for major productivity losses in the near future. To prevent these losses, research needs to identify factors that can improve employees’ psychological health.

Previous research has provided inconsistent results regarding the relationship between different leadership styles such as transformational leadership, consideration, initiating structure and employees’ psychological health [1, 39, 43]. In addition, only a few studies have examined the relationship of leadership styles that have been proposed to promote socially responsible and moral behaviors to employees’ psychological health [30]. In the present study, we aim to address this gap in leadership research by examining the relationship between servant leadership and employees’ psychological health.

Servant leadership is a leadership style that focuses on serving multiple stakeholders of the organization. Hale and Fields [26] define servant leadership as “an understanding and practice of leadership that places the good of those led over the self-interest of the leader, emphasizing leader behaviors that focus on follower development, and de-emphasizing glorification of the leader” (p. 397). On the basis of this definition, we predict that the proposed beneficial relationships between servant leadership and health should become manifest in negative relationships between servant leadership and short- and long-term indicators of psychological strain, which are thought to reflect overall employees’ psychological health. These predictions draw on organizational fit theory [15–17] and social identity theory [48]. Organizational fit theory proposes that psychological health arises from a high fit between employees’ needs and organizational provisions and affordances. According to social identity theory, individuals define themselves as members of groups, and thereby internalize social identities that serve to structure both organizational perceptions and behavior. Amongst other things, Haslam et al. [27] argue that shared social identity serves as a basis for feelings of trust, support, and belongingness—feelings that in turn are expected to improve employees’ psychological health. In the present paper, we integrate both theories and propose that servant leaders can create a shared social identity among followers and thus fulfill followers’ needs. Consequently, we expect servant leadership to be positively related to health because it speaks to followers’ needs for a sense of shared social identity and hence provides a high needs-supply fit.

In the present study, we examine the proposed relationships in regard to long and short-term indicators of job-strain (ego depletion, need for recovery, emotional exhaustion and depersonalization) which reflect employees’ psychological health. Furthermore, we simultaneously test the relationship of servant leadership and job ambiguity a well-known job stressor, which has been repeatedly found to predict strain. In this way, we may provide evidence that servant leadership accounts for unique amounts of variance in indicators of strain over and above that explained by job stressors such as job ambiguity.

We believe that our research has the capacity to provide several contributions to the literature on leadership and health. First, it may provide initial evidence about the nature of the relationship between servant leadership and various indicators of psychological health. Second, it examines whether servant leadership may explain unique parts of variance in indicators employees’ psychological health over and above job ambiguity a major job-stressor.

In the following, we will first elaborate on the construct of servant leadership and distinguish it from other leadership styles. Next, we will focus on the relationship between servant leadership and employees’ psychological health, and present a theoretical foundation for the proposed beneficial relationships. Finally, we will present the details of our analyses.

1.1 Servant Leadership

The previously described characteristics of servant leadership such as behaving ethically, helping followers grow and succeed, putting followers first [35] reflect that servant leadership focuses on multiple stakeholders of the organization such as shareholders, the community, customers, and especially followers. The core idea of servant leadership is that managers set aside their personal self-interest for the benefit of collective interests [9, 23]. Thus, servant leaders do not lead for their own benefit or for the benefit of their organizations, but to integrate the interests of multiple stakeholders, and especially their employees. Consequently, servant leaders do not lead through formal authority or charisma as other leadership styles such as transformational leadership [13, 52] suggest, but instead rely on “one-on-one” communication to understand the abilities, needs, desires, goals and the potential of their employees [35, p. 162]. Additionally, servant leaders shape their employees’ views and values to encourage them to become servants and servant leaders themselves [25].

Even though servant leadership overlaps with other leadership styles such as transformational, ethical and authentic leadership [5, 11, 51], it also differs from these constructs in certain key aspects. First, Graham [24] argues that leadership styles such as transformational leadership fail to consider the importance of a moral compass, which constitutes a crucial aspect of servant leadership. Second, in contrast to leadership styles that include ethical aspects (e.g., authentic leadership and ethical leadership), servant leadership focuses on the success of multiple stakeholders of the organization. Third, servant leadership is especially focused on the interests, and competencies of followers. Thus, servant leaders aim to develop their employees, and to support their growth and success [46].

Recent research provides strong support for the idea that servant leadership exerts unique beneficial effects on various job attitudes, fairness perceptions, and also job performance. For example, Ehrhart [22] demonstrated that servant leadership accounts for additional variance in commitment (5 %), job satisfaction (7 %), perceived supervisor support (4 %) and procedural justice (8 %) over and above leader member exchange (LMX) and transformational leadership. These results are also supported by further research that provided evidence for beneficial effects of servant leadership on employee work outcomes over and above other leadership styles (e.g., transformational leadership, and LMX) [21, 35, 40, 44]. In conclusion, servant leadership is characterized by unique behavioral patterns and attitudinal aspects, which are distinct from other related leadership concepts and thus account for a broad spectrum of positive outcomes, even after controlling for other aspects of leadership. Yet despite its various beneficial outcomes, to the best of our knowledge, no previous studies have examined the relationship between servant leadership and employees’ psychological health.

1.2 Servant Leadership as a Predictor of Employees’ Psychological Health

As previously described the relationship between servant leadership and employees’ psychological health is grounded in principles of organizational fit theory [15–17] and social identity theory [48]. On the dyadic level, we argue that servant leaders create work environments that have affordances that fulfill employees’ individual needs. According to organizational fit theory, a high fit between a person and its environment reduces strain while a low fit is expected to increase strain. In this regard, previous fit research has identified two types of person-environment fit. The first type is commonly referred to as demands-abilities fit. It describes whether situational demands can be met by a person’s abilities. The second type is referred to as needs-supply fit and relates to the match between a person’s needs and provisions, and affordances in a given environment. We expect that the beneficial relationship between servant leadership and employees’ health results primarily from a high degree of needs-supply fit.

At the same time, on a group level, we argue that a shared social identity influences employees’ psychological health by creating an atmosphere of trust, support, justice, and belongingness [27]. Social identity theory suggests that in various contexts individuals define themselves as members of social groups (e.g., as ‘us’ family, friends, and colleagues). These social groups provide individuals with personal security, and emotional bonding. At the same time though, individuals tend to experience negative psychological consequences if they lack or lose social identity (e.g., if they are rejected from groups) and positive consequences if they maintain or gain a sense of shared social identity (e.g., if they identify with groups [15] see also Cruwys et al. [20]). The core aspects of servant leadership speak to issues of social identity. For example, by promoting ethical behaviors among their followers, servant leaders establish norms that are embraced by all members of their group, and followers’ enactment of these norms helps to establish a shared sense of positive social identity. At the same time, followers’ sense that leaders are ‘doing it for us’ (rather than for themselves) should help to cultivate both followership and psychological health [28].

Integrating both theories, we argue that through its focus on helping followers grow and achieve agency by acting in a manner consistent with commonly agreed views and values, servant leaders establish a shared social identity among group members. Thereby, servant leaders shape their followers’ needs and help to create work-environments that fulfill these needs. For example, a servant leader may emphasize the importance of giving back to the community and induce the need to be involved in community service or volunteer activities. Then again, the servant leader may fulfill this need by introducing community service activities at corporate events. In short, we argue that the beneficial relationship between servant leadership and employees’ psychological health reflects processes occurring at both group and dyadic levels. At the group level, it arises from feelings of trust, support, and belongingness that result from the leader’s cultivation of a sense of shared social identity; at the dyadic level, it results from the leader’s enhancement of needs-supply fit.

1.3 The Present Research

We conducted a study to provide empirical evidence for the proposed beneficial relationships between servant leadership and employee’s psychological health operationalized as indicators of strain. To demonstrate that servant leadership shares unique proportions of variance with the examined indicators of strain, we simultaneously tested servant leadership in combination with job ambiguity a well-known job stressor. Job ambiguity involves a perceived lack of job-related information and reflects employees’ perceptions of uncertainty concerning various aspects of their jobs [10]. The negative impact of job ambiguity on, for example, burnout has been observed in multiple studies. Schwab and Iwanicki [45] reported that among teachers, role conflict and job ambiguity account for considerable proportions of variance in emotional exhaustion (23 %) and depersonalization (20 %). Furthermore, in their meta-analysis, Lee and Ashforth [34] found moderate to strong correlations of job ambiguity to both burnout symptoms.

Furthermore, we test the relationship of servant leadership with different indicators of strain. We examine emotional exhaustion and depersonalization as long-term indicators of strain. Emotional exhaustion is considered to be the main component of burnout and is defined as a chronic state of depletion and fatigue resulting from one’s work [36, 53]. Depersonalization, another dimension of burnout, refers to negative and cynical attitudes towards people at work [15]. As short-term indicators of strain, we examine ego depletion and need for recovery. Ego depletion refers to a momentary state of regulatory resource depletion following regulatory demands [7]. According to resource-based conceptualizations of psychological strain [38], ego depletion can be thought as a short-term correlate of impairments in psychological health. Need for recovery reflects the need to recuperate from work tasks that is strongest in the last hours of work and directly after work [50].

In sum, we test the impact of servant leadership on employee health (a) on both long- and short-term indicators of strain and (b) competing for variance with a well-known stressor. We argue that the proposed approach should allow us to test whether the relationship between servant leadership and employees’ health is invariant to different boundary conditions such as individual indicators of strain, and whether servant leadership explains unique amounts of variance in employees’ psychological health. Thus, our hypothesis is:

Servant leadership is beneficially related to psychological health over and above job ambiguity. In view of the examined indicators of strain servant leadership is negatively related to (a) ego depletion, (b) need for recovery, (c) emotional exhaustion and (d) depersonalization.

2 Method

2.1 Participants

The sample for our study was recruited from a major bank in Germany. In their work environment, job ambiguity constitutes one of the main stressors because employees have to balance customers’ needs with organizational priorities [18]. For example, even though employees need to sell financial products to customers, they also have to consider customers’ interests. Thus, we decided to control for job ambiguity as a stressor in this study.

All participants were contacted via e-mail and received an online survey which was completed during regular working hours. Participation was voluntary, and all participants were assured that their responses would remain confidential. Out of 705 contacted persons, we received 443 responses (63 % response rate). Participants’ age ranged from 18 to 60 years (M = 39.22; SD = 10.68). Of these participants, 56 % were female, and 23 % worked part-time.

2.2 Measures

Job ambiguity was assessed with nine items from a scale developed by Breaugh and Colihan [10], which was translated and validated in German by Sodenkamp and Schmidt [47]. Here participants indicated their perceived lack of job-related information on a 7-point Likert-scale (1 = not at all, 7 = a great deal). Sample items are “I know how to get my work done (what procedures to use)” (work method ambiguity) and “I know when I should be doing a particular aspect (part) of my job” (scheduling ambiguity). All items were recoded so that higher scores reflect greater job ambiguity and were then averaged to a single scale score.

We used Ehrhart’s [22] scale to measure servant leadership. On the basis of a literature review, Ehrhart [22] identified seven major categories of servant leader behaviors (forming relationships with followers, empowering followers, helping followers grow and succeed, behaving ethically, having conceptual skills, putting followers first, and creating value for those outside of the organization). Afterwards, he developed two items for each category resulting in a 14-item measure of servant leadership. Factor analyses revealed that this measure had a one-dimensional structure [22, 49]. For the present study, the items were translated into German through a three-step procedure. This involved the original items being translated into German, then back into English, and then compared. In our study, participants rated the behavior of their leader on these items using 5-point Likert-scales (1 = not at all, 5 = a great deal). Table 1 gives an overview of the original items and the German translations.

Table 1 Servant leadership: Original items and German translations
Full size table

The measurement of ego-depletion was based on five items that assessed the current experience of resource depletion and low will-power (e.g., “At the moment, I feel increasingly less able to focus on something.” or “At the moment, I feel as if I have no willpower left”). The scale was originally developed and validated by Bertrams et al. [8] and all items were scored on 5-point Likert-scales (1 = not at all, 5 = a great deal). Day-specific need for recovery was assessed with five items (e.g., “After the present day’s work I feel so tired that I cannot get involved in other activities.” or “My job causes me to feel rather exhausted.” [50]). All items were scored on the same 5-point Likert-scales.

The burnout dimensions of emotional exhaustion (eight items) and depersonalization (five items) were assessed with a German version [14] of the Maslach and Jackson [36] Burnout Inventory. Sample items are “I feel emotionally drained from my work” (emotional exhaustion) and “I have become more callous toward people since I took this job” (depersonalization). The items were rated on a 6-point Likert-scale (1 = not at all, 6 = very strong).

2.3 Factor Structure

Before testing our hypotheses, we examined whether the factor structure of our translated measurement of servant leadership resembles the factor structure of the original measurement. Thus, we conducted an exploratory factor analysis (EFA). Results of the EFA (principal component analysis with orthogonal varimax rotation) provided a two-factor structure of servant leadership. The amount of variance explained by both factors was 62.7 %. The first factor includes the first 12 Items (cf. Table 1). These items reflect the prioritization of subordinates concerns. The first factor accounts for the largest proportion of variance (53.7 %). The factor loadings range from .83 to .51. Items 13 and 14 represent the second factor. This factor reflects encouragement of ethical and prosocial behavior. It accounts for 8.9 % of variance. The factor loadings range from .73 to .67. Because Costello and Osborne [19] argue that factors with less than three items tend to be unstable, we conducted another EFA and specified the extraction of only one factor as suggested by the original scale [22]. This factor accounts for 53.7 % of the total variance. The factor loadings ranged between .83 and .51. Because the one-factor model resembled the original scale and the factor loadings did not fall below the critical value of .30 [19], we decided to use a composite measure of servant leadership. Thus, as suggested by Ehrhart [22], this involved averaging responses to all 14 servant leadership items (cf. Table 1).

2.4 Data Analysis

Before testing the hypotheses, we conducted Harman’s one-factor test [42] in order to analyse potential confounding effects due to common method variance. The results of this test suggested that a common method factor accounted for 31.3 % of variance. On this basis, we infer that our results are not seriously biased by high common method variance. Subsequently, we analysed our data using three-step hierarchical linear regression analyses with ego depletion, need for recovery, emotional exhaustion and depersonalization as outcomes. In Step 1, we entered the control variables of age, gender, and working time into the regression to control for their potential confounding influences on the relationships under examination [31, 37]. In Step 2, we introduced job ambiguity. In Step 3, servant leadership was added into the regressions.

3 Results

Table 2 presents the means, standard deviations, and reliabilities of study variables.

Table 2 Means, standard deviations, internal consistencies (Cronbach’s alpha) and intercorrelations
Full size table

Our hypothesis proposed that servant leadership is negatively related to ego depletion, need for recovery emotional exhaustion and depersonalization over and above job ambiguity. Results of multiple regression analyses relating to this hypothesis are presented in Table 3. These indicate that, after controlling for demographic variables, job ambiguity is positively related to ego depletion, need for recovery (cf. Table 3), emotional exhaustion and depersonalization (cf. Table 4). Moreover and theoretically more important, servant leadership was negatively related to all four outcomes (ego depletion [β = −.15; p < .01], need for recovery [β = −.18; p < .01], emotional exhaustion [β = −.25; p < .01] and depersonalization [β = −.25; p < .01]). The incremental variance explained by servant leadership was respectively 2 % for ego depletion, 2 % for need for recovery 6 % for emotional exhaustion and 5 % for depersonalization (cf. Table 3). These results thus support our hypothesis.

Table 3 Regression results (β values) for ego depletion and need for recovery
Full size table
Table 4 Regression results (β values) for emotional exhaustion and depersonalization
Full size table

4 Discussion

The aim of the present research was to provide evidence for the hypothesized beneficial relationships between servant leadership and employees’ psychological health. We demonstrated that servant leadership accounts for additional variance in short- and long-term indicators of strain (ego depletion, need for recovery, emotional exhaustion and depersonalization) over and above job ambiguity, a well-known job stressor. These results support our claims that servant leadership is (a) beneficially related to short- and long-term indicators of psychological health and (b) that it accounts for unique variance in these indicators of psychological health over and above job-specific stressors.

4.1 Theoretical Implications

Our study provides several important contributions to research on servant leadership. First, we know of no previous research that has examined the relationship between servant leadership and employees’ psychological health. Integrating our results and previous research on servant leadership demonstrates that servant leadership does not only improve outcomes such as job attitudes and job performance [49], but also that these improvements may also have benefits for employees’ health. This is especially important because research on other leadership styles such as transformational leadership indicates that these leadership styles are primarily beneficial for the organization or the leader regardless of potential negative consequences for employees [37].

Second, our findings demonstrate that servant leadership accounts for variance in indicators of employees’ psychological health over and above job ambiguity as a prominent job-stressor. These results indicate that the effects of servant leadership are unique.

4.2 Limitations and Suggestions for Further Research

It remains the case, however, that our study also has several limitations that need to be discussed. First, our study variables were all operationalized by means of self-report. Thus, common method variance or a self-report bias might have contaminated the observed relationships [42]. However, Harman’s one-factor test indicated that common method influences were not a severe biasing factor in our data.

Second, although we assumed a particular causal order of the variables, the correlational design of our research does not permit causal conclusions. Other causal directions or even reciprocal relations could be possible as well. For example, psychologically healthy employees might have made the choice to work for servant leaders. Additionally, as we suggested in our theoretical argument, shared social identity might have been a third factor that influenced both the perception of servant leadership on the part of employees as well as indicators of employees’ psychological health. While we cannot rule out this possibility, research on alternative leadership styles and health has also demonstrated the health enhancing effects of leadership [39]. Additionally, given that we controlled for the effects of job ambiguity (as a strong predictor of psychological health) on the relationship between servant leadership and psychological health, alternative causal paths seem rather unlikely. However, further research should focus on disentangling the relationship between servant leadership and psychological health. In particular, it is necessary to examine more closely the mechanisms (e.g., shared social identity, needs-supply fit [16, 27]) that may account for the beneficial relationship between servant leadership and employee health.

4.3 Practical Implications

The results of our study indicate that servant leadership is beneficially related to employees’ psychological health. Thus, it appears that organizations that seek to improve the psychological health of employees should consider encouraging their leaders to lead on the basis of the principles of servant leadership. This can be achieved through leadership training and through the role modeling of servant leadership by current leaders.

More specifically, in the first instance, practitioners might design leadership training programs that elaborate on the basic principles of servant leadership such as forming relationships with followers, empowering followers, helping followers grow and succeed, and behaving ethically. Part of such training involves discussing different ethical perspectives and reflecting on how a servant leader might engage constructively with these perspectives. Another useful exercise may be to simulate and record annual appraisals of employees and evaluate these appraisals according to principles of servant leadership (e.g., appreciation of employees). Additionally, in this training program, trainees may also learn about the benefits of servant leadership over and above other leadership styles.

Second, research on servant leadership argues that this leadership style can be best taught by setting examples [4]. Thus, being led by a servant leader is expected to increase the likelihood of a follower becoming a servant leader. Accordingly, leaders should provide examples and lead in accordance with the principles of servant leadership with a view to encourage employees to follow in their footsteps. Indeed, in so far as the present evidence suggests, broad acceptance of servant leadership will have positive consequences for employees’ psychological health. Thus, it appears that there is much to be gained through the transmission of this approach from one generation of leaders to the next.