Considering the orphan: Personal identification and its relations with transformational leadership,trust,and performance in a three-path mediation model

Abstract

Transformational leadership research investigating the identification processes of followers has mostly focused on social identification. In contrast, empirical evidence on the role of followers’ personal identification with their leader is still scarce. Furthermore, there has been no attempt to test the links between personal identification and other proposed mediators of transformational leadership. This study examined how a group’s identification with and trust in its leader sequentially combine to mediate the relationship between transformational leadership and supervisor-rated group performance. Using a sample of 343 employees and their supervisors (N = 80), we tested a model in which trust functions as a proximal mediator and personal identification as a distal mediator in the leadership–performance relationship. Results support the hypothesized three-path mediation model. Our findings indicate the importance of trust for the development of personal identification in the leadership process and provide evidence that the group’s identification with the leader results in positive outcomes for organizations.

Keywords

Transformational leadership personal identification trust group performance three-path mediation model

Introduction

Over the last decade, leadership research has mainly focused on transformational leadership as one of the most popular leadership paradigms of the last 50 years (Judge and Piccolo, 2004). Transformational leadership (Bass, 1985) includes the following dimensions: idealized influence attributed, idealized influence behavior, inspirational motivation, intellectual stimulation, and individualized consideration. Scholars around the world have given evidence of the positive effects of transformational leadership on followers’ motivation and performance (e.g. Judge and Piccolo, 2004). However, a proper understanding of leadership effectiveness also requires comprehension of the psychological processes mediating the influence of a leader’s behaviors on the reactions of his or her followers. Specifically, scholars (e.g. van Knippenberg et al., 2004; Walumbwa and Hartnell, 2011) have called for more research on multiple mediators and their interrelations in the transformational leadership process. To date, only a few studies have analyzed multiple mediators sequentially. Some exceptions include the studies by Walumbwa and Hartnell (2011) on self-efficacy and relational identification, by Pillai et al. (1999) on procedural justice and trust, and by Piccolo and Colquitt (2006) on core job characteristics as first mediator and goal commitment and intrinsic motivation as second mediators.

While theories of transformational and charismatic leadership differ in some aspects, all of them identify personal identification with the leader, defined as the extent of inclusion of the leader in the subordinates’ self (Aron, 2003; van Knippenberg et al., 2004), and the creation of trust as essential elements of the transformational leader–follower relationship (Bass, 1985; Conger and Kanungo, 1998; Shamir et al., 1993). Neither transformational leadership theory nor research, however, provide clear answers on how these psychological processes are interrelated and intertwined in the transformational leadership process. In contrast to the well-defined role of trust (Dirks and Ferrin, 2002), empirical evidence is still scarce on how followers’ personal identification¹ (Aron, 2003; van Knippenberg et al., 2004) is related to transformational leadership and potential outcomes. Only recently have scholars begun to analyze the role of personal identification in the transformational leadership process (Kark et al., 2003; Liu et al., 2010; Zhu et al., 2013). Otherwise, the concept has remained chronically underinvestigated (van Knippenberg et al., 2004) and has been “treated like an orphan” in the identity and self-construal literature: Scholars have mainly focused on social identification with a collective and paid only little attention to personal identification (van Knippenberg et al., 2004). Considering the importance of leader–follower relationships for the leadership process (Graen and Uhl-Bien, 1995), more evidence concerning the role of self-construal on the relational level is clearly needed (Kark et al., 2003; Liu et al., 2010; van Knippenberg et al., 2004).

Further, with regard to the consequences of personal identification in the transformational leadership process, empirical findings are somewhat mixed. A number of scholars have stated that the potential “dark side” of charisma results in dependency (Kark et al., 2003) and unquestioning conformity (e.g. Howell and Shamir, 2005). This may be explained by identifying with the leader, revealing a negative effect of personal identification. However, as followers’ personal identification with the leader should also be associated with their motivation to fulfill the goals and vision of the leader (e.g. Shamir et al., 1993), positive performance outcomes are just as likely. As initial findings of Liu et al. (2010) and Zhu et al. (2013) suggest, strong personal identification with the leader also encourages followers to speak up (e.g. share critical thinking with the leader) rather than keep silent, enhances organizational commitment and innovativeness, and reduces their turnover intentions, thus demonstrating positive effects of personal identification on follower behavior.

By considering the combined effect of trust and personal identification in a more comprehensive three-path mediation model, we aim to contribute to the literature in two important ways. First, our study attempts to extend the existing body of knowledge on transformational leadership by considering the interrelationship of different mediators in the transformational leadership process. Specifically, we draw upon transformational models of trust (Lewicki and Bunker, 1995) to explain how transformational leadership enhances a group’s identification with a leader by way of trust-building behavior. Second, we seek to explain how transformational leadership relates to group performance through the groups’ trust and identification with the leader, giving further evidence of the beneficial effects of high identification in the leadership process. In this way, we add to the understanding of personal identification and its consequences and aim to reduce the imbalance of research on identification processes (i.e. social versus personal identification) in the leadership literature. To our knowledge, this is the first study to examine the combined effect of trust and personal identification on performance and their relation in the transformational leadership process to date.

Level of conceptualization of transformational leadership and outcomes

As leadership occurs in a broader context, like a work group or unit, we adopted the perspective that leadership can be conceptualized as a group-level phenomenon. Leaders may communicate the same mission and goals, stress the same values, and build trust among all group members. Thus, group members share the same leadership “environment” that, in turn, sets the stage for similar perceptions among group members regarding the leader and the relationship to the leader (Jung and Sosik, 2002). Our approach is in line with research on transformational leadership (e.g. Jung and Sosik, 2002; Kark et al., 2003; Shamir et al., 1998) that suggests that leaders act transformationally upon whole groups or units, instead of directing their behavior toward specific individuals. Employees working in the same group under the same leader are likely to be influenced by similar leadership behaviors (Lord et al., 1999; Shamir et al., 1998) which should therefore be operated on a group level (Purvanova et al., 2006).

In this study, we also view the reactions of the followers to transformational leadership (trust in the leader and identification with the leader) as group phenomenon in terms of the consensus perceptions of the members, analogous to macroperceptions in climate research. In support of our approach, aggregating group members’ individual perceptions of transformational leadership and trust in the leader to the group level (i.e. averaging their ratings) is common practice (e.g. Jung and Sosik, 2002; Kark et al., 2003; Shamir et al., 1998), particularly when these group perceptions are linked to group-level outcomes, such as group performance (e.g. Chen et al., 2007; Dirks, 2000). Although identification with the leader has not been analyzed as consensus perception of group members yet, we believe that it may be conceptualized and interpreted at a group level, as well. Importantly, Katz and Kahn (1978) stated in their open system theory that phenomena on the individual and group level concerning leadership, motivation, and performance are strongly connected. Translated into methodological terms, this view is congruent with a direct consensus composition model (Chan, 1998), assuming that a construct that has been conceptualized and operationalized at the individual level is functionally equivalent to the construct at the group level. That means that the consensus perception among group members represents the meaning of the group construct, thereby constituting functional equivalence at the two levels. Functional equivalence of relationships at different organizational levels has been supported by multilevel studies showing the same pattern of relations for the individual and the group level, for example for leadership, empowerment, and performance (e.g. Chen et al., 2007).

Our conceptualization of transformational leadership, trust, and identification with the leader as group perceptions is congruent to the focal unit of the study, the group. This is particularly important as the dependent variable, group performance, may be regarded as a function of the perceptions of the whole group (Dirks, 2000). In contrast to group perceptions, the group’s performance is a group-level variable in the first place and was measured with reference to the group rather than the individual by the respective leader.

In the following, we will review transformational leadership theory and research in order to provide a rational for the proposed relationships. In doing so, we apply a functional equivalence perspective (Chan, 1998) assuming that followers’ individual perceptions of transformational leadership, trust in the leader, and identification with the leader will be functionally equivalent when conceptualized and transferred to the group level in terms of consensus group perceptions. Therefore, we assume that the same theory (e.g. Bass, 1985) and findings conceptualized and found for the individual-level may be validly applied to the group level (i.e. to relationships between consensus perceptions), as well. The proposed research model is depicted in Figure 1.

Figure 1.

The hypothesized research model.

Transformational leadership and followers’ trust

Transformational leadership theories name a number of leader behaviors that contribute to the creation of trust, defined as the “the extent to which a person is confident in, and willing to act on the basis of, the words, actions, and decisions of another” (McAllister, 1995: 25). Likewise, a group’s trust in the leader may be conceptualized as an “expectation or belief that the team can rely on the leader’s actions or words and that the leader has good intentions toward the team” (Dirks, 2000: 1004). As leaders have the most formal power on their work groups, trust in the leader is an essential aspect for groups’ functioning and well-being (Dirks, 2000). Transformational leaders demonstrate their trustworthiness to the group by serving as a role model showing exemplary behavior and communicating important values and goals which they themselves live up to (e.g. Bass, 1985). A meta-analysis by Dirks and Ferrin (2002) has pointed to the importance of trust in the transformational leadership process, showing that transformational leadership had the largest relationship with trust out of all observed variables (r = .72). Therefore, we propose:

Hypothesis 1: A group’s perception of transformational leadership is positively associated with the group’s trust in its leader.

Transformational leadership, followers’ trust, and personal identification

A follower’s self-concept, defined as all the knowledge one has about oneself, spans an individual, relational, and collective level, each representing different aspects of a person’s identity (Brewer and Gardner, 1996). Self-construal on the relational level—that is, personal identification—refers to the extent to which individuals define themselves in terms of a significant other, experiencing a strong sense of connectedness with the other person (van Knippenberg et al., 2004). According to Aron (2003), when personally identifying with another person, cognitive representations of the self and of the other overlap and become part of one’s own self-representations.

Several scholars (Bass, 1985; Conger and Kanungo, 1998; Shamir et al., 1993; van Knippenberg et al., 2004) have proposed a relationship between transformational leadership and personal identification with the leader. As the leader is one of the most influential others for followers in an organizational context, he or she may represent a significant source of a follower’s self-construal or identity (Shamir et al., 1993). When followers identify with their transformational leader, followers regard their leader as an important representative component of their self or identity and incorporate the leader’s vision, values, and goals as their own guiding principles (Shamir et al., 1993).

Importantly, identification with the leader is conceptualized in a broader sense as and goes beyond the mere internalization or adoption of the leader’s vision and goals. Whereas internalization refers to the process of integration and acceptance of opinions, norms, and values that have been established by influential others because of the content of these attitudes (Kelman, 1961), identification involves the integration of the significant other into one’s self-concept or identity because the other person is self-defining (one wants to be like or gain the qualities of the other) and satisfies the individual’s underlying needs for self-enhancement, self-expansion, and meaning (Ashforth and Mael, 1989; Kelman, 1961). Following the above conceptualizations, a group’s identification with its leader may be defined as the collective sense of connectedness with the leader as an important part of group members’ identity and the collective adoption of the leader’s vision, values, and goals as guiding principles.

Transformational leaders are presumed to raise the intrinsic value of efforts and goals by linking followers’ identities to the leader’s mission, which then become self-referential for followers (e.g. Shamir et al., 1993; van Knippenberg et al., 2004). This is achieved by the enthusiastic communication of an inspirational and emotionally appealing vision while emphasizing the ideological importance of the mission and instilling confidence and optimism in the group that the challenging goals can be achieved (e.g. Shamir et al., 1993). As a consequence, task attainment becomes a way for the group to express their identity: followers view their job as meaningful as it serves the greater good and helps to accomplish the leader’s mission (Shamir et al., 1993). Further, due to the transformational leader’s extraordinary appeal and unconventional behavior, group members identify with the leader who is seen as a role model, want to imitate and please the leader, and seek for his or her approval in order to enhance their self-confidence and self-worth (Conger and Kanungo, 1998; Shamir et al., 1993). In sum, groups may identify with their transformational leader as this satisfies group members’ needs for self-enhancement and meaning. Several studies have confirmed the positive relationship between transformational leadership and identification with the leader (Kark et al., 2003; Liu et al., 2010; Zhu et al., 2013), though the research which analyzes mediators of the transformational leadership–personal identification link is still scarce (van Knippenberg et al., 2004).

So how are trust in and identification with the leader related in the transformational leadership process? As an initial hint, Yukl (1989) stated that trust is the basis for followers’ willingness to be influenced by their transformational leader and may therefore be an important antecedent for adopting and integrating the leader’s values and an inner merging. Bass and Riggio (2006) claim that confidence and trust in a leader strengthen followers’ identification with their leader, and this is an important driver of the followers’ commitment to performance goals and outcomes. The assumption that high levels of trust may be the basis for personal identification in the transformational leadership process has not been investigated empirically, yet. However, a relationship between trust and personal identification has previously been found in other leadership contexts, for example in studies on distanced leadership, i.e. leadership of geographically dispersed teams (Connaughton and Daly, 2004).

As transformational leadership theories do not explicitly state whether and how trust in and identification with the leader may be related to each other, we draw upon transformational models of trust (e.g. Lewicki and Bunker, 1995) that illustrate how trust may strengthen personal identification with the leader in the transformational leadership processes. These models suggest that trust is a multifaceted construct consisting of different qualitative stages that are defined in terms of the basis of trust. The idea of trustworthiness presented in the transformational leadership literature (Bass, 1985; Conger and Kanungo, 1998; Shamir et al., 1993) may be represented by the two lower levels of trust, that is, calculus-based trust (CBT) and knowledge-based trust (KBT) (Lewicki and Bunker, 1995). CBT is conceptualized as an evaluation of the costs and benefits resulting from creating and sustaining a relationship versus ending it. KBT, however, is grounded in the knowledge and understanding of the relationship partner and the resulting predictability of his or her behavior (Lewicki and Bunker, 1995).

As transformational leaders clearly explicate what behaviors on the part of the followers will be rewarded while acting as they say, followers are able to calculate expected costs and benefits and discover whether CBT may be justified. The consistency and congruence between the leaders’ communicated values and actions create a sense of credibility (Gabarro, 1987). As the leader takes personal risk to achieve the vision and demonstrates courage and conviction, group members accumulate personal knowledge about their leader as being competent and skilled. According to Gabarro (1987) and Butler (1991), perceived competence is one of the most important elements of trust. Further, the leader’s demonstration of concern and care for the group’s needs contributes significantly to group members’ trust as their confidence in the leader’s goodwill and intentions grows (Barber, 1983; Bass, 1985). Followers’ perception of their leader’s exemplary behavior, competence, and consideration lead to KBT, the next level of trust.

CBT and KBT build the foundation for identification-based trust, the highest form of trust defined in terms of shared values, the internalization of goals and preferences, and the identification with the other (Aron, 2003; Lewicki and Bunker, 1995; McAllister, 1995).When the transformational leader is seen as trustworthy and faithful (CBT and KBT), positive emotional bonds between followers and the transformational leader may evolve through mutual reliance and support. As group members feel valued and supported by the trusted transformational leader who shows care and concern for their development and needs (Bass, 1985; Shamir et al., 1993) while taking personal risks and making self-sacrifices (Conger and Kanungo, 1998), the group develops admiration for their leader, resulting in the attribution of charisma and the identification with their leader (Conger and Kanungo, 1998). As the transformational leader acts as a trusted role model and provides the group with meaning and confidence, the group understands, shares, and internalizes the expectations and goals of the trusted leader and, as a consequence, strives to act for the leader and his or her appealing vision. Therefore, we propose:

Hypothesis 2: A group’s trust in its leader mediates the relationship between the group’s perception of transformational leadership and the group’s identification with its leader.

Followers’ personal identification and group performance

According to the self-concept-based theory of charismatic leadership by Shamir et al. (1993), identification with the leader is crucial for understanding the beneficial effects of transformational leadership on the followers’ (group) performance. As stated by Andersen and Chen (2002), identification with a leader may motivate followers to act in accordance to the leader’s vision and values because identification with a leader creates a sense of oneness, so that goals and interests of the leader are seen as one’s own. As a consequence, followers may act for the benefit of the leader showing high performance, because this is, to some extent, equivalent to acting for their own benefit. Further, because the group’s performance and task attainment has been linked by the transformational leader to a meaningful mission, group members perceive their work as being more intrinsically motivating, resulting in high effort and performance (Shamir et al., 1993). Thus, when adopting their transformational leader’s high performance standards and goals, the group will show extra effort and will be motivated to perform well (Bass, 1985) as the “behavioral manifestations of such a commitment are likely to further bind the self-concept of the individual to the leader and the mission” (Shamir et al., 1993: 584). This effect is further enhanced because followers strive for their leader’s recognition and approval while trying to avoid feelings of guilt due to failure. They work hard and show extra effort in order to maintain the self-defining and rewarding relationship (Conger and Kanungo, 1998; Shamir et al., 1993).

As identification with the leader is one of the key mechanisms of transformational leadership effectiveness mentioned in common leadership theories, it is particularly surprising that research on the influence of identification with the leader on performance outcomes is still scarce. However, Zhu et al. (2013) could show that identification with the leader enhances followers’ innovativeness and Walumbwa and Hartnell (2011) recently found a positive relationship for relational identification (that focuses on the role relationship with the leader instead of the leader itself) and followers’ performance via self-efficacy. Based on the theoretical arguments and preliminary empirical findings, we propose:

Hypothesis 3: A group’s identification with its leader will be positively associated with group performance.

Followers’ trust, personal identification, and group performance

The important role of leadership behavior for the creation and preservation of trust, and the central function of interpersonal trust for the effectiveness of a team have been recognized by theory and research (Bass, 1985; Dirks, 2000; Dirks and Ferrin, 2002). Trust in one’s leader is of great importance for functioning at a group level as it sets the basis for the group’s willingness to accept the leader’s mission, goals, and decisions and to work hard to achieve them (Dirks, 2000). Podsakoff et al. (1990) suggested that followers’ trust in a leader is one of the most important mediators of the transformational leadership–performance relationship. This has been confirmed by a number of studies (Casimir et al., 2006; Jung and Avolio, 2000; Podsakoff et al., 1990). Findings of the meta-analysis by Dirks and Ferrin (2002) indicated a relatively small but significant relationship between trust and job performance (r = .16).

Based on transformational models of trust (Lewicki and Bunker, 1995), we argue that the transformational leader’s creation of a trusting relationship with his or her group results in the group members’ identification with its leader (Lewicki and Bunker, 1995; McAllister, 1995), and this translates trust into high group performance. When trust in a leader (CBT and KBT) is established within a group through the leader’s personal sacrifice, concern, and commitment to the mission (Bass, 1985; Shamir et al., 1993), followers eventually identify with their leader and his or her goals, needs, and values. This inner merging and the integration of the leader’s values and goals may explain how a trusting relationship may result in followers’ commitment to performance goals and outcomes (Bass and Riggio, 2006): The group’s desire to please the trusted leader and the adoption of the leader’s high performance goals and vision (perceived as their own) give direction and orientation to followers’ trust-based willingness to devoted acting (e.g. Dirks, 2000; Shamir et al., 1993). This willingness and intrinsic motivation to work for the mission of the leader that is perceived as one’s own explains why the group may perform beyond expectation (Bass, 1985). In other words, the trusted transformational leader links the group’s sense of identity to his or her own vision and values, thereby fostering high group performance. Thus, we propose:

Hypothesis 4: A group’s identification with its leader mediates the relationship between the group’s trust in its leader and group performance.

Three-path mediation model

Past studies investigated either the mediating role of trust or the mediating role of personal identification with the leader in the transformational leadership process. One exception is the study by Shamir et al. (1998) which addressed both trust in a leader and identification with a leader simultaneously but merged both concepts into one measure. Thus, the question of how trust and personal identification with a leader are interrelated in their function as mediators of the effectiveness of transformational leadership remains to be tested. As a conclusion of all the above presented arguments, in line with transformational leadership theory and transformational trust models, we propose a three-path mediation model in which transformational leadership is related to group members’ trust, which, in turn, is linked to groups’ identification with their leader that, again, is related to group performance:

Hypothesis 5: The influence of a group’s perception of transformational leadership on group performance will be mediated by the group’s trust in its leader as a first mediator, and the group’s identification with its leader as a second mediator in a three-path mediation model.

A critical evaluation of the above proposed relationships may bring up the question of whether the reversed sequence or the simultaneous functioning of the two mediators (trust and personal identification with a leader) may be possible or likely, as well. Personal identification with a leader may also build the basis for followers’ trust in the leader in the transformational leadership process. Accordingly, some authors (Gillespie and Mann, 2004; Jones and George, 1998) point to the critical role of sharing or identifying with the leader’s values for the establishment of trust in the leader. Conversely, Sitkin and Roth (1993) have shown that perceptions of value incongruence can quickly lead to distrust. Further, it might also be suggested that trust in and identification with a leader simultaneously mediate the influence of transformational leadership on performance as they may be linked in a reciprocal way. Barber (1983) stated, for example, that shared values help to create a relationship built on trust and that trust, in turn, is needed to express and maintain the particular shared values from which trust originates. Thus, in order to provide further evidence for the robustness and validity of our hypotheses, we will test whether the hypothesized model depicted in Figure 1 will show a better fit to our data than a three-path mediation model with the reversed order of mediators and a two-path mediation model with the mediators specified simultaneously.

Method

Sample and procedure

To recruit study participants, we approached different companies in diverse industries and the nonprofit sector from all over Germany. The study was introduced as research on “leadership behavior and development,” and we emphasized the voluntary nature and confidentiality of the participants’ responses. To enhance participation, managers were given detailed individual feedback on their leadership behavior and the option for a personal consulting in exchange for cooperation. In total, the final sample comprised 343 employees and 80 managers at 23 organizations (32.1% service industry, 35.8% manufacturing industry, 32.1% nonprofit organizations). Managers were asked to distribute questionnaires to all of their team members. Participation was voluntary for employees, as well. Managers received instructions for their employees orally as well as general information about the research, its purpose for leadership development, and the data collection procedure. Employees were asked to send back their questionnaires in enclosed envelopes to the second author, preserving anonymity.

Managers and their employees were from different functional backgrounds, such as accounting and controlling, human resources, marketing and customer services, production, and research and development. Group sizes varied between 2 and 13 with an average of 4.29 (SD = 1.89). Data on the control, independent, and mediating variables were collected from the employees, and data on the dependent variable were collected from their supervisors; 39.2% of the employee sample were female, the average age was 42.19 years (SD = 5.81), and the average organizational tenure and group membership were 11.26 years (SD = 8.79) and 5.84 years (SD = 4.48), respectively. The sample of managers included lower (32.9%), middle (40.5%), and top management (26.6 %); 23.8% was female and the average age was 43.62 years (SD = 6.67). The average tenure in the organization was 11.25 years (SD = 6.90), and average managerial experience was 10.27 years (SD = 6.71).

Measures

Transformational leadership

Employees were asked to rate their supervisors according to the extent to which they agreed with the statements about their transformational leadership behavior. Since this study was conducted within the German organizational context, we used 24 items of a modified German version of the Multifactor Leadership Questionnaire (Bass and Avolio, 1995), translated by Felfe and Goihl (2002) to measure transformational leadership. This adapted version has been empirically validated (Felfe, 2006) and widely used in research on transformational leadership in Germany (e.g. Heinitz et al., 2005). Transformational leadership was measured with the five scales: idealized influence attributed, idealized influence behavior, inspirational motivation, intellectual stimulation, and individualized consideration, as well as an additional sixth scale, charisma. As the single components of transformational leadership usually show high intercorrelations (Bass and Avolio, 1995) and in accordance with theory and research (Bass, 1999; Jung and Sosik, 2002; Kark et al., 2003), we conceptualized transformational leadership as a one-dimensional concept (α = .94). Ratings were completed on a five-point scale ranging from 1 = not at all to 5 = frequently, if not always.

Personal identification

Following Kark et al. (2003), identification with the leader was measured with six items that were adopted from the identification scale developed by Mael and Ashforth (1992). The items focused on the manager who supervised the work group to which the respondent belonged. A sample item is “When someone praises my supervisor, it feels like a personal compliment.” Responses were made on a five-point scale ranging from 1 = strongly disagree to 5 = strongly agree.

Trust

Referring to Conger and Kanungo (1998: study 6), trust in one’s leader was measured by items taken from Bass (1985) and Butler (1991). The items were “I have complete faith in him/her” (Bass, 1985) and “I can count on him/her to be trustworthy” (Butler, 1991). This scale was also anchored on a five-point response scale ranging from 1 = strongly disagree to 5 = strongly agree.

Supervisor-rated group performance

Managers were asked to use 9 items to assess their employees’ group performance on a six-point scale (ranging from 1 = not fulfilled (0%) to 6 = clearly exceeded (>150%)). The items referred to overall goal attainment for efficiency of work flows, profits, sales, productivity, cost reduction, customer satisfaction, team climate, absence rate, and turnover. Teams and their managers did have control over most of the performance indicators as reflected in managers’ formal power over/responsibility for the mentioned outcomes.² Items represented general performance indicators that are widely referred to in the management (Kaplan and Norton, 1992) and scientific (Rauch et al., 2000) literature. These general performance indicators were identified as being included in most of the participating companies’ performance evaluation systems, thus, assuring appropriateness of these indicators for managers’ performance ratings in this study. Thereby, leaders were actually able to refer to the latest performance rating results of their team, so that the data had a high practical relevance and content validity. By using criteria well established in management practice, we wanted to ensure managers’ acceptance of and participation in the study. This was more likely as managers could relate their own leadership feedback to criteria actually used in their own company. Internal validity of this scale was good (α = .82).

Control variables

We controlled for managers’ and employees’ gender, employees’ length of work group membership, and leader’s leadership experience (both measured by number of years) based on past research (Siders et al., 2001) and because of their correlation pattern with the relevant constructs (p < .10) in all the analyses described below. As employee age and group tenure were strongly correlated and showed a similar pattern of relations with the outcome variables, we decided to control for employees’ group tenure only in order to avoid redundancy of controls and because group tenure was more strongly related to the outcomes.

Construct validity

Prior to the test of our hypotheses, we analyzed the adequacy and distinctness of the employee self-report measures (Anderson and Gerbing, 1988). We tested for common method variance by conducting a confirmatory factor analysis using AMOS 18 (Arbuckle, 2009). To assess whether each of the scales (for transformational leadership) and each of the measurement items (for trust and personal identification) would load significantly onto the construct or scales with which they are associated, the fit of a three-factor model was compared with the fit of two alternative, more constrained, models: a single-factor model in which measures of all three factors were set to load on a single factor and a two-factor model in which measures of trust and personal identification were set to load on one factor. Results showed (see Table 1) that the hypothesized three-factor model fit the data significantly better than the alternative single-factor model (χ² (3) = 620.69, p = .000) and two-factor model (χ² (1) = 298.59, p = .000). Thus, results indicate validity and independence of our measures (Anderson and Gerbing, 1988).

Table 1.

Results of a confirmatory factor analysis including transformational leadership, trust, and personal identification.

Measurement models	χ²	df	p	GFI	CFI	TLI	RMSEA	RMSEA 90% CI
One-factor model	865.376	77	.000	.681	.724	.674	.173	(.163; .184)
Two-factor model	543.278	75	.000	.833	.836	.801	.135	(.125; .146)
Three-factor model	244.686	74	.000	.905	.940	.927	.082	(.071; .094)

GFI: goodness-of-fit index; CFI: comparative fit index; TLI: Tucker–Lewis index; RMSEA: root mean square error of approximation.

Level of analysis

In this work, we focused on the consensus perceptions and responses among group members. We recognize, however, that there is some debate whether aggregating individual level constructs (i.e. transformational leadership perceptions, identification with and trust in the leader) to the group level is meaningful (Chan, 1998; Chen et al., 2004; Mathieu and Chen, 2011). Because the focal referent of follower ratings belonging to one work group was the same (i.e. the leader of that work group), it seems quite logical to transfer the idea of transformational leadership as a group phenomenon to follower responses to that certain leadership behavior. Thus, we argue that the mediating mechanisms linking the group members’ consensus perception of transformational leadership and their performance may be similar for followers of the same leader, too. We propose that a similar or consensus perception of transformational leadership may result in equally homogenous psychological processes (i.e. similar levels of trust and identification) and to collective or joint behavioral reactions (group performance) among followers of the same leader (e.g. Jung and Avolio, 1999; Jung and Sosik, 2002; Keller, 2006). Therefore, we aggregated group members’ perceptions of transformational leadership, trust in the leader, and identification with the leader to the group level. Group members’ performance (as observable construct) was assessed via supervisor ratings of the work group as a whole.

Analysis strategy

Drawing on past research and recommendations, we followed the structural equation modeling (SEM) approach and used path analyses (Bollen, 1998) for testing the proposed relationships. We tested for mediation through a set of nested model comparisons, as has been recommended in the literature (e.g. Holmbeck, 1997). In this way, we attempted to determine if full mediation or partial mediation resulted in the best fit within the model comparisons. In all our analyses, control variables were included by using residualized variables for both mediators and outcome. For an evaluation of the model fit and the fit comparisons to nested models, we used several goodness-of-fit indices, such as the chi-square statistic divided by the degrees of freedom (χ²/df), the goodness-of-fit index (GFI), the comparative fit index (CFI), the Tucker–Lewis index (TLI), and the root mean square error of approximation (RMSEA). We applied as acceptable goodness-of-fit criteria a χ²/df ratio smaller than 3, GFI, CFI, and TLI values greater than .90, and a RMSEA lower or equal to .08 (Hu and Bentler, 1998; Kline, 2005).

Although analyses of mediation chains longer than two paths (i.e. with more than one mediator) are still relatively rare in organizational behavior research, they are becoming more common (cf. Taylor et al., 2008). This also accounts for many organizational theories and models proposing two or more mediators between their constructs. As was suggested by MacKinnon and colleagues (MacKinnon et al., 2002; Taylor et al., 2008), we used the joint significance test for testing the three-path mediated effects. The joint significance test by MacKinnon et al. (2002) is based on James and Brett’s (1984) definition of mediation. In contrast to the classical Baron and Kenny (1986) approach, it does not require the overall relation between the predictor and the outcome to be significant. According to Taylor et al. (2008), the joint significance test may easily be generalized to the context of three-path mediation models: mediation is evident if each of the three paths in the mediated effect is significantly nonzero. The simultaneous test of the significance of the three paths (the test SEM applies) represents a suitable method because of its simplicity and controlling of Type I error at or below its nominal level, as well as its good power (MacKinnon et al., 2002; Taylor et al., 2008).

Results

Aggregation analyses

As described above, aggregating group members’ perceptions of transformational leadership as well as associated constructs (e.g. trust in the leader) at a group level is a widely used procedure (e.g. Dirks, 2000; Jung and Sosik, 2002; Kark et al., 2003). In order to determine if we were justified in analyzing our hypotheses at the group level, we tested whether aggregation was valid for transformational leadership and the mediating variables trust and personal identification with the leader (Bliese, 2000). As recommended by LeBreton and Senter (2008), we computed both r_wg values to assess within-group agreement (James et al., 1993) as well as intra-class correlation statistics (ICC[1] and ICC[2]; Bliese, 2000) for assessment of between-group reliability. The r_wg as a measure of consensus presumes that consensus truly reflects the group as a unit (cf. Guzzo et al., 1993). As noted by Kozlowski and Klein (2000), the sharedness of perceptions is a key factor for the determination of the appropriate level of analysis. Significant ICC(1) values indicate that analyses on the individual level are inappropriate due to dependence in the data structure; ICC(2) values are an indicator for between-group differences. For group members’ perception of transformational leadership, the median r_wg value was .98, the average ICC(1) value was .24, and the average ICC(2) value was .58 (F = 2.37, p < .001). For group members’ identification with the leader, the median r_wg value was .89 and the average ICC(1) and ICC(2) values were .08 and .27, respectively (F = 1.39, p < .01). Group members’ trust in the leader had a median r_wg value of .86, an average ICC(1) value of .26, and an average ICC(2) value of .60 (F = 2.48, p < .001). Although there are no strict cut-offs for determining if aggregation based on r_wg and ICC values is appropriate, an r_wg equal to or greater than .70 and ICC(1) values exceeding .05 (Bliese, 2000) are considered sufficient to justify aggregation (see e.g. Avolio et al., 2004). Furthermore, Glick (1985) underlines the importance of a significant ANOVA (F-value) indicating significant between-group variance. As F-values were significant for all variables and our analyses revealed values higher than those reported in organizational research dealing with aggregation issues (Bliese, 2000), we concluded that the aggregation of group members’ perceptions of transformational leadership behaviors, identification with and trust in the leader was statistically suitable. Thus, we subsequently tested our hypotheses and report on relationships at the group level of analysis (N = 80).

Descriptive analyses

Table 2 presents means, standard deviations, alpha coefficients, and intercorrelations for all study variables. Reliabilities (Cronbach’s alpha) were between .82 and .94 and can be considered good, overall.

Table 2.

Means, standard deviations, reliabilities (Cronbach’s alpha), and intercorrelations among study variables.

Variable	M	SD	1	2	3	4	5	6	7	8	9	10
1. Age (employees)	42.19	5.81	–
2. Gender (employees)	.40	.34	−.04	–
3. Group membership (years)	5.85	4.50	.55***	.00	–
4. Age (leader)	43.62	6.67	.43***	−.10	.30**	–
5. Gender (leader)	.24	.43	−.18	.30**	−.14	−.05	–
6. Leadership experience (years)	10.27	6.71	.45***	−.22	.35**	.69***	−.13	–
7. Transformational leadership	3.62	.43	.01	−.45***	−.08	.07	−.22	.20	(.94)
8. Trust in the leader	4.00	.46	−.05	−.35**	−.15	−.03	−.13	−.03	.72***	(.88)
9. Personal identification	2.75	.44	−.16	−.34**	−.23*	−.12	−.11	.04	.44***	.50***	(.82)
10. Group performance	3.71	.64	−.27*	−.06	−.29**	−.12	.03	−.09	.13	.09	.29*	(.82)

N = 80. Reliability coefficients appear in parentheses. Gender: 0 = male; 1 = female.

p < .05, **p < .01, ***p < .001 (two-tailed).

Hypotheses testing and model comparisons

First, we run several path analyses using AMOS 18 (Arbuckle, 2009) to test a set of nested model comparisons (e.g. Holmbeck, 1997) in order to reveal whether full mediation or partial mediation (direct, proximal, unconstrained model, see below) was most adequate for our research model (for a comparison of these models, see Figure 2). Second, in order to test our hypotheses, we inspected the direct and indirect effects of the best-fitting model resulting from these nested model comparisons, as suggested by Taylor et al. (2008) and in line with other scholars (e.g. Judge and Colquitt, 2004). Finally, to give a more rigorous test of our hypotheses, we conducted some further analyses comparing our research model to alternative (nonnested) models in which the position of the two mediators was changed (for a comparison of these models, see Figure 3).

Figure 2.

Comparison of nested mediation models for the hypothesized research model.

Figure 3.

Comparison of nonnested mediation models with different sequences.

First, in order to determine the nature of the proposed mediation effect, we compared

A full mediation model with the following alternative partial mediation models with (an) additional path(s), as depicted in Figure 2.

A direct model in which the group’s perception of transformational leadership has also direct effects on the groups’ identification with the leader and group performance.

A proximal model in which the group’s perception of transformational leadership has an additional direct effect on the group’s identification with the leader and the group’s trust has an additional direct effect on group performance.

An unconstrained model in which the group’s perception of transformational leadership and the group’s trust both have (additional) direct effects on the group’s identification with the leader and group performance, respectively.

Results revealed that the model indicating full mediation fit the data best. As shown in Table 3, the direct, proximal, and unconstraint models provided poorer fits in comparison to the full mediation model. Further evidence is provided by the nonsignificant differences in chi-square between the full and the alternative mediation models. Including additional paths from transformational leadership to identification with the leader and performance (direct model) did not significantly improve model fit (Δχ²[2] = 1.00, p = .61). Similarly, neither the inclusion of paths from transformational leadership to identification with the leader and from trust to performance in the proximal model (Δχ²[2] = .45, p = .80) nor the inclusion of paths from transformational leadership and trust to performance and from transformational leadership to personal identification in the unconstraint model significantly improved model fit (Δχ²[3] = 2.59, p = .46). Thus, none of the additional paths add to the explanation of the construct variance. We retained the fully mediated, most parsimonious version, as the preferred model.

Table 3.

Testing the three-path mediation model: Comparisons of nested path models.

Measurement models	χ²	df	p	GFI	CFI	TLI	RMSEA	RMSEA 90% CI
Full mediation	2.592	3	.459	.984	1.000	1.014	.000	(.000; .180)
Direct model	1.593	1	.207	.990	.990	.937	.087	(.000; .327)
Proximal model	2.137	1	.144	.987	.980	.880	.120	(.000; .350)
Unconstraint model	.000	0	1.000	1.000

Results are for an omnibus test where two mediators are specified simultaneously in a model.

GFI: goodness-of-fit index; CFI: comparative fit index; TLI: Tucker–Lewis index; RMSEA: root mean square error of approximation.

Second, we tested our hypotheses by further investigating the direct and indirect effects of the best-fitting full mediation model. Concerning the main effects (depicted in Figure 4), the proposed relationship between transformational leadership and trust (Hypothesis 1) and the relation between identification with a leader and performance (Hypothesis 3) were both significant. Our hypothesis concerning mediation would be supported if the simultaneous test of the paths in the mediated effect (i.e. the indirect effects, presented in the footnote of Figure 4) would be significant (Taylor et al., 2008). Hypothesis 2 predicted that the relationship between transformational leadership and personal identification would be mediated by trust. As the significant indirect effect of transformational leadership on personal identification indicates, the effect of leadership on identification was transmitted through trust, thereby supporting Hypothesis 2. As shown by the significant indirect effect of trust on performance, the mediation of the relationship between trust and performance by personal identification (Hypothesis 4) was supported by the data as well. Finally, the significant indirect effect of transformational leadership on group performance confirms a sequential mediation model in which the effect of transformational leadership on group performance is transmitted via the group’s trust as a first mediator and the groups’ identification with the leader as a second mediator (Hypothesis 5).

Figure 4.

Results of the three-path full mediation Model A. The path coefficients and squared multiple correlations are standardized. Standardized indirect effects: Transformational leadership → Personal identification = .26**; Trust → Group performance = .12*; Transformational leadership → Group performance = .08*. *p < .05, **p < .01 (2-tailed).

Finally, in order to rule out alternative explanations and to further test the robustness of our hypothesized model (Model A), we compared this model with two alternative models (e.g. Carmeli et al., 2009), as depicted in Figure 3. Thus, we tested Model B in which the sequence of the two mediators was reversed and Model C in which the two mediators were entered in the path analysis together (i.e. at the same position). We compared these three models (A, B, and C) by consulting Akaike’s information criterion (AIC) and the Bayesian information criterion (BIC). The AIC and BIC belong to the class of predictive fit indices that are used to select among competing nonhierarchical models estimated with the same data. The model with the lowest AIC and BIC values is chosen as the one most likely to replicate, i.e. with a relatively better fit and fewer parameters compared with competing models (Kline, 2005). As shown in Table 4, the model with the lowest AIC and BIC value is Model A with trust as a proximal and personal identification with the leader as a distal mediator. Thus, Model A shows the best model fit and is the one most likely to replicate compared to Model B and Model C. In sum, Hypothesis 5 proposing that the leadership–performance relationship is mediated by trust as a first (proximal) mediator and personal identification as a second (distal) mediator can be further supported by the results, proving the robustness of Model A.

Table 4.

Testing three alternative full mediation models: Comparisons of Models A, B, and C.

Measurement models	χ²	df	p	GFI	CFI	TLI	RMSEA	RMSEA 90% CI	AIC	BIC
Model A	2.592	3	.459	.984	1.000	1.014	.000	(.000; .180)	16.592	33.266
Model B	39.560	3	.000	.831	.355	−.290	.393	(.289; .506)	53.560	70.234
Model C	2.137	1	.144	.987	.980	.880	.120	(.241; .363)	20.137	41.575

Results are for an omnibus test where two mediators are specified simultaneously in a model.

GFI: goodness-of-fit index; CFI: comparative fit index; TLI: Tucker–Lewis index; RMSEA: root mean square error of approximation; AIC: Akaike’s information criterion; BIC: Bayesian information criterion.

Discussion

In the present study, we seek to shed light on the roles of group members’ trust in and identification with their leader in their function as mediators in the transformational leadership process when both variables are analyzed in combination. To our knowledge, no prior study has examined the joint effect of these explanatory mechanisms, analyzing how they are intertwined when predicting group performance as a distal outcome of transformational leadership.

The most important finding of this study is the identification of a three-path mediated model in which the group members’ trust in their leader fully mediates the effect of transformational leadership on a groups’ identification with its leader, and identification, in turn, fully mediates the relationship between trust and the performance of the group. A group that perceives its leader as transformational will place more trust in him or her, which in turn leads to higher identification with its leader. The identification with the leader, that is, group members’ consensus perception of being one with their leader, having the same goals and values, will lead to higher performance of the group. This is in line with the well-established finding that trust in a leader is strongly related to transformational leadership (Dirks and Ferrin, 2002), working as a key mechanism of transformational leadership effectiveness. Although past research has examined the role of trust and identification processes in explaining the relationship between transformational leadership and work-related outcomes, we are not aware of a study that has sequentially analyzed trust and identification with a leader as mediating mechanisms, nor are we aware of any study analyzing identification with the leader as a group phenomenon. By specifying the relationship between the proposed (e.g. Bass, 1985; Shamir et al., 1993) but only individually tested (e.g. Kark et al., 2003; Podsakoff et al., 1990) mediators of transformational leadership (trust and personal identification) as well as a three-path mediation model, we shed some further light on the inner workings of transformational leadership (Bass, 1999).

Future studies should specify how different combinations of mediators are related and intertwined in the transformational leadership process, thus clarifying which are most effective in predicting relevant work-related outcomes. A promising methodological approach of future (transformational) leadership research may be the investigation of additional three-path mediated models (Taylor et al., 2008). Specifically, the psychological processes described in the self-concept theory by Shamir et al. (1993) explicitly propose sequential effects of transformational leadership on self-concept-related, motivational, and behavioral follower outcomes. One such sequential effect has been tested by Walumbwa and Hartnell (2011) who show that transformational leaders impact their followers’ performance by enhancing the self-efficacy of their followers and, in turn, relational identification. Piccolo and Colquitt (2006) demonstrated that the beneficial effect of transformational leadership on followers’ task performance and OCB can be explained by followers’ perception of their job as enriching that, in turn, enhances their intrinsic motivation and goal commitment. This line of research may also include the identification of potential moderators that may influence the associations among constructs, that is, moderated mediation and mediated moderation (e.g. Walumbwa et al., 2008). As pointed out by van Knippenberg et al. (2004), followers’ self-concepts, such as collective or individual efficacy or self-esteem, likely function as moderating factors diminishing or enhancing the effect of transformational leadership on the attitudes and behaviors of followers. Research has shown, for example, that followers’ efficacy referring to available resources (their means efficacy) enhances the mediating effect of identification with the work unit and thus linking transformational leadership and employee performance (Walumbwa et al., 2008). Further, it might also be interesting to analyze other forms or foci of identification as has been proposed by Ashforth et al. (2008) who investigated the identification with different roles.

Our study extends previous research on followers’ self-construal by focusing on the role of followers’ personal identification with their leader instead of their social identification with the organization or work unit as a predictor for work group performance. To date, identification with a leader has been an underinvestigated concept in transformational leadership research (van Knippenberg et al., 2004). The results provide evidence to the proposed but untested function of personal identification with the leader as a mediator of the leadership–performance relationship. In line with some initial evidence (Liu et al., 2010; Zhu et al., 2013), our results demonstrate that a group’s identification with its leader may lead to beneficial outcomes for the group in terms of enhanced performance. However, as some scholars (Kark et al., 2003) have shown that identification with the transformational leader may also lead to negative outcomes, such as greater dependency of followers on their leader, future research has to identify whether the positive effect of followers’ identification on their performance may also be explained by greater dependency. Even though followers show higher loyalty and willingness to make an effort for their leaders’ goals and visions when personal identification is high, Howell and Shamir (2005) point out that followers may be less likely to critically evaluate and question the leader’s mission and goals. Thus, overidentification, “blind faith”, and dependency may be a problem when personal identification with the leader is high, reflecting the “dark” influence of personalized transformational relationships. As personalized relationships are more likely when followers’ self-evaluations and self-concept clarity is low (Howell and Shamir, 2005), future studies should include follower characteristics, such as self-esteem, self-efficacy, and self-concept clarity, as moderating variables in the transformational leadership process.

Practical implications

The findings of this study allow for a number of implications for organizational and leadership practice. First, they point to the importance of leader–follower relationships in organizations and may lead to a deeper understanding of the role of consensus attitudes and evaluations within a group in the transformational leadership–performance link. Looking at the recent corporate scandals and managerial frauds that have been vividly discussed in the public and taking into account that followers’ identification with the leader involves the internalization of the leader’s values and goals, it seems especially important that the leader’s values suit the ethical conduct of the organization and serve its overall goals. As Howell and Shamir (2005) have noted, relationships between transformational leaders and their followers may be characterized as either personalized or socialized. In personalized relations, in which the transformational leader is motivated by the need to accumulate personal power instead of striving for the collective good, identification with the leader may have negative effects for followers and organizations. Therefore, it seems important to select and develop leaders who act as role models, display ethical behavior, and promote goals that contribute to the good of the organization. Managers should be encouraged to treat followers in a respectful and caring way, foster positive trustful relationships, and refrain from self-serving behavior.

Second, the knowledge of the specific mediating variables linking leadership behavior and important work-related outcomes may be helpful in the conceptualization of leadership development and training. Transformational leadership behaviors emphasizing the creation of their group members’ trust and identification, such as role modeling and individualized consideration, should be particularly focused and trained. Leaders could also learn about the role that consistency of words and actions plays, and the importance of living up to the values and goals communicated to their followers. Further, demonstrating competence and effort to reach the organizational goals could be addressed in coaching or training. Thereby, leaders could learn how trust-building and identification-building behaviors may increase their group’s performance.

Limitations and future research

The present study has several limitations that are important to note. One major weakness of our study is the use of a cross-sectional design, which does not allow for clear causal conclusions. The posited relations should therefore additionally be analyzed in longitudinal studies and experimental settings. Further, although the groups’ consensus perception of transformational leadership, trust, and identification with their leader had—in the logic of the constructs—to be assessed by followers’ self-reports, common-source bias may be a problem (Podsakoff et al., 2003). However, we tried to reduce common-source bias by assessing followers’ performance through ratings of their supervisors. Although the results of a single-factor test indicated that common-method bias may not be a problem and Chan (2009) suggested that problems associated with self-report data may be an “urban legend” and overstated in the literature, this potential bias has to be kept in mind when interpreting the results. Finally, it might be noteworthy that relationships between transformational leadership, trust, and personal identification are fairly reasonable and as high as could be expected from past research (Dirks and Ferrin, 2002; Kark et al., 2003).

Another constraint is that the influence of important contextual factors on group members’ trust (in the leader), such as the influence of fairness perceptions and organizational justice (De Cremer et al., 2007), have not been included in the present study. De Cremer et al. (2007) found that interactional justice most dominantly affected followers’ transformational leadership perceptions. Thus, an alternative explanation that has yet to be tested is that the interrelations between transformational leadership, trust, and personal identification with one’s leader were produced by the leader’s interpersonally just behavior. Also, follower characteristics like personality constructs (Nübold et al., 2013) or need configurations (Wofford et al., 2001) that have not been controlled for in this study might have influenced followers’ perception of or reactions to transformational leadership. Concerning personality constructs, McCrae and Costa (1987) have already stated that individuals high in neuroticism and antagonism are less likely to be trusting. Likewise, past theory and research on trust (Dirks and Ferrin, 2002) have demonstrated that individuals differ in their readiness to trust another person, conceptualizing trust as a personality characteristic of the trustee. Future research should therefore also explore the hypothesized and confirmed model in the context of follower-centered approaches of leadership focusing on follower or group characteristics as important influencing factors of leadership effectiveness.

A further constraint of the present study is that we only focused on German companies and organizations, hence, the generalizability of the effects to other cultures and countries may not be justified (e.g. Jung and Avolio, 2000). In order to maximize generalizability, future research should investigate the proposed model in other cultures as well. Finally, we analyzed only one type of leadership, namely transformational leadership. Other types of leadership that also incorporate trust as an important outcome or mediator, such as leader–member exchange (LMX, Graen and Uhl-Bien, 1995) or ethical leadership (Brown et al., 2005), may or may not show the same pattern of relationships with trust, personal identification with the leader, and performance. Specifically, one could argue that followers who are members of the leader’s in-group will be more trusting than members of the out-group as there is a lower level of mutual influence between leader and followers in the out-group (Podsakoff et al., 1990). Therefore, generalizability to other leadership models has yet to be tested in future studies.

Conclusions

The aim of our study was to develop a more complete understanding of the inner workings of transformational leadership (Bass, 1999) and the role of groups’ identification with their leader in the transformational leadership process (van Knippenberg et al., 2004). More research is needed in the area of explanatory mechanisms of transformational leadership effectiveness and their interconnections among one another. Future studies should identify other joint mediation effects that contribute to a successful cooperation between transformational leaders and their work group. The results of this study show that the knowledge of the means through which transformational leadership achieves high group performance helps to improve leadership theory and research, as well as organizational practice.

Footnotes

Notes

Author biographies

Annika Nübold is a post-doctoral Researcher and Lecturer at the Department of Work and Organizational Psychology at Bielefeld University, Germany. In her research, she focuses on personality dynamics at work, specifically core self-evaluations, leadership and the role of ethics in organizations.

Stefan L Dörr is Consultant for Management Diagnostic and Leadership Development and is Business Partner at A47 Consulting in Munich, Germany. He has worked for several years as Professor for Work and Organizational Psychology. His research is mainly concerned with leader motives and transformational leadership.

Günter W Maier is Professor for Work and Organizational Psychology at Bielefeld University, Germany. His major research interests are personality at work, leadership, organizational justice, innovative behavior, and personal work goals. He has several years of experience in personnel selection, training and consulting.

References

Andersen

Chen

(2002) The relational self: An interpersonal social-cognitive theory. Psychological Review 109(4): 619645.

Anderson

Gerbing

(1988) Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin 103(3): 411423.

Arbuckle

(2009) Amos 18 UserE´s Guide, Crawfordville, FL: Amos Development Corporation.

Aron

(2003) Self and close relationships. In: Leary

Tangney

(eds) Handbook of Self and Identity, New York, NY: The Guilford Press, pp. 442461.

Ashforth

Harrison

Corley

(2008) Identification in organizations: An examination of four fundamental questions. Journal of Management 34(3): 325374.

Ashforth

Mael

(1989) Social identity theory and the organization. Academy of Management Review 14(1): 2039.

Avolio

Zhu

Koh

(2004) Transformational leadership and organizational commitment: Mediating role of psychological empowerment and moderating role of structural distance. Journal of Organizational Behavior 25(8): 951968.

Barber

(1983) The Logic and Limits of Trust, New Brunswick, NJ: Rutgers University Press.

Baron

Kenny

(1986) The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology 51(6): 11731182.

10.

Bass

(1985) Leadership and Performance beyond Expectations, New York, NY: Free Press.

11.

Bass

(1999) Two decades of research and development in transformational leadership. European Journal of Work and Organizational Psychology 8(1): 932.

12.

Bass

Avolio

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

13.

Bass

Riggio

(2006) Transformational Leadership, Mahwah, NJ: Lawrence Erlbaum.

14.

Bliese

(2000) Within-group agreement, non-independence, and reliability: Implications for data aggregation and analyses. In: Klein

Kozlowski

SWJ

(eds) Multilevel Theory, Research, and Methods in Organizations: Foundations, Extensions, and New Directions, San Francisco, CA: Jossey-Bass, pp. 349852.

15.

Bollen

(1998) Path analysis. In: Armitage

Colton

(eds) Encyclopedia of Biostatistics, Sussex, UK: John Wiley, pp. 32803284.

16.

Brewer

Gardner

(1996) Who is this “we”? Levels of collective identity and self representations. Journal of Personality and Social Psychology 71(1): 8393.

17.

Brown

Treviño

Harrison

(2005) Ethical leadership: A social learning perspective for construct development and testing. Organizational Behavior and Human Decision Processes 97(2): 117134.

18.

Butler

(1991) Toward understanding and measuring conditions of trust: Evolution of a conditions of trust inventory. Journal of Management 17(3): 643663.

19.

Carmeli

Ben-Hador

Waldman

(2009) How leaders cultivate social capital and nurture employee vigor: Implications for job performance. Journal of Applied Psychology 94(6): 15531561.

20.

Casimir

Waldman

Bartram

(2006) Trust and the relationship between leadership and follower performance: Opening the black box in Australia and China. Journal of Leadership and Organizational Studies 12(3): 6884.

21.

Chan

(1998) Functional relations among constructs in the same content domain at different levels of analysis: A typology of composition models. Journal of Applied Psychology 83(2): 234246.

22.

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, NY: Routledge, pp. 309335.

23.

Chen

Kirkman

Kanfer

(2007) A multilevel study of leadership, empowerment, and performance in teams. Journal of Applied Psychology 92(2): 331346.

24.

Chen

Mathieu

Bliese

(2004) A framework for conducting multilevel construct validation. In: Yammarino

Dansereau

(eds) Research in Multilevel Issues: Multilevel Issues in Organizational Behavior and Processes, Oxford: Elsevier, pp. 273303.

25.

Conger

Kanungo

(1998) Charismatic Leadership in Organizations, Thousand Oaks, CA: Sage.

26.

Connaughton

Daly

(2004) Identification with leader: A comparison of perceptions of identification among geographically dispersed and co-located teams. Corporate Communications: An International Journal 9(2): 89103.

27.

De Cremer

van Dijke

Bos

(2007) When leaders are seen as transformational: The effects of organizational justice. Journal of Applied Social Psychology 37(8): 17971816.

28.

Dirks

(2000) Trust in leadership and team performance: Evidence from NCAA basketball. Journal of Applied Psychology 85(6): 10041012.

29.

Dirks

Ferrin

(2002) Trust in leadership: Meta-analytic findings and implications for research and practice. Journal of Applied Psychology 87(4): 611628.

30.

Enders

(2001) A primer on maximum likelihood algorithms available for use with missing data. Structural Equation Modeling 8(1): 128141.

31.

Felfe J (2006) Validierung einer deutschen Version des “Multifactor Leadership Questionnaire” (MLQ Form 5 x Short) von Bass und Avolio (1995) [Validation of a German version of the “Multifactor Leadership Questionnaire” (MLQ 5X–Short) by Bass and Avolio (1995)]. Zeitschrift für Arbeits- und Organisationspsychologie 50(2): 61–78.

32.

Felfe

Goihl

(2002) Deutsche überarbeitete und ergänzte Version des “Multifactor Leadership Questionnaire” (MLQ) [German revised and supplemented version of the “Multifactor Leadership Questionnaire” (MLQ)]. In: Glöckner-Rist

(eds) ZUMA-Informationssystem. Elektronisches Handbuch Sozialwissenschaftlicher Erhebungsinstrumente. Version 5.00, Mannheim: Zentrum für Umfragen, Methoden und Analysen.

33.

Gabarro

(1987) The Dynamics of Taking Charge, Boston, MA: Harvard Business School Press.

34.

Gillespie

Mann

(2004) Transformational leadership and shared values: The building blocks of trust. Journal of Managerial Psychology 19(6): 588607.

35.

Glick

(1985) Conceptualizing and measuring organizational and psychological climate: Pitfalls in multilevel research. Academy of Management Review 10(3): 601616.

36.

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(2): 219247.

37.

Guzzo

Yost

Campbell

(1993) Potency in groups: Articulating a construct. British Journal of Social Psychology 32(1): 87106.

38.

Heinitz

Liepmann

Felfe

(2005) Examining the factor structure of the MLQ: Recommendation for a reduced set of factors. European Journal of Psychological Assessment 21(3): 182190.

39.

Holmbeck

(1997) Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the child-clinical and pediatric psychology literatures. Journal of Consulting and Clinical Psychology 65(4): 599610.

40.

Howell

Shamir

(2005) The role of followers in the charismatic leadership process: Relationships and their consequences. Academy of Management Review 30(1): 96112.

41.

Bentler

(1998) Fit indices in covariance structure modeling: Sensitivity to underparameterized model misspecification. Psychological Methods 3(4): 424453.

42.

James

Brett

(1984) Mediators, moderators, and tests for mediation. Journal of Applied Psychology 69(2): 307321.

43.

James

Demaree

Wolf

(1993) rwg: An assessment of within-group interrater agreement. Journal of Applied Psychology 78(2): 306309.

44.

Jones

George

(1998) The experience and evolution of trust: Implications for cooperation and teamwork. Academy of Management Review 23(3): 531546.

45.

Judge

Colquitt

(2004) Organizational justice and stress: The mediating role of work-family conflict. Journal of Applied Psychology 89(3): 395404.

46.

Judge

Piccolo

(2004) Transformational and transactional leadership: A meta-analytic test of their relative validity. Journal of Applied Psychology 89(5): 755768.

47.

Jung

Avolio

(1999) Effects of leadership style and followers' cultural orientation on performance in group and individual task conditions. Academy of Management Journal 42(2): 208218.

48.

Jung

Avolio

(2000) Opening the black box: An experimental investigation of the mediating effects of trust and value congruence on transformational and transactional leadership. Journal of Organizational Behavior 21(8): 949964.

49.

Jung

Sosik

(2002) Transformational leadership in work groups: The role of empowerment, cohesiveness, and collective efficacy on perceived group performance. Small Group Research 33(3): 313336.

50.

Kaplan

Norton

(1992) The balanced scorecard – measures that drive performance. Harvard Business Review 70(1): 7179.

51.

Kark

Shamir

Chen

(2003) The two faces of transformational leadership: Empowerment and dependency. Journal of Applied Psychology 88(2): 246255.

52.

Katz

Kahn

(1978) The Social Psychology of Organizations, New York, NY: Wiley.

53.

Keller

(2006) Transformational leadership, initiating structure, and substitutes for leadership: A longitudinal study of research and development project team performance. Journal of Applied Psychology 91(1): 202210.

54.

Kelman

(1961) Processes of opinion change. The Public Opinion Quarterly 25(1): 5778.

55.

Kline

(2005) Principles and Practice of Structural Equation Modeling, New York, NY: Guilford Press.

56.

Kozlowski

SWJ

Klein

(2000) A multilevel approach to theory and research in organizations: Contextual, temporal, and emergent processes. In: Klein

Kozlowski

SWJ

(eds) Multilevel Theory, Research, and Methods in Organizations: Foundations, Extensions, and New Directions, San Francisco, CA: Jossey-Bass, pp. 390.

57.

LeBreton

Senter

(2008) Answers to 20 questions about interrater reliability and interrater agreement. Organizational Research Methods 11(4): 815852.

58.

Lewicki

Bunker

(1995) Trust in relationships: A model of development and decline. In: Bunker

Rubin

(eds) Conflict, Cooperation, and Justice: Essays Inspired by the Work of Morton Deutsch, San Francisco, CA: Jossey-Bass Publishers, pp. 133173.

59.

Liu

Zhu

Yang

(2010) I warn you because I like you: Voice behavior, employee identifications, and transformational leadership. Leadership Quarterly 21(1): 189202.

60.

Lord

Brown

Freiberg

(1999) Understanding the dynamics of leadership: The role of follower self-concepts in the leader/follower relationship. Organizational Behavior and Human Decision Processes 78(3): 167203.

61.

McAllister

(1995) Affect- and cognition-based trust as foundations for interpersonal cooperation in organizations. Academy of Management Journal 38(1): 2459.

62.

McCrae

Costa

(1987) Validation of the five-factor model of personality across instruments and observers. Journal of Personality and Social Psychology 52(1): 8190.

63.

MacKinnon

Lockwood

Hoffman

(2002) A comparison of methods to test mediation and other intervening variable effects. Psychological Methods 7(1): 83104.

64.

Mael

Ashforth

(1992) Alumni and their alma mater: A partial test of the reformulated model of organizational identification. Journal of Organizational Behavior 13(2): 103123.

65.

Mathieu

Chen

(2011) The etiology of the multilevel paradigm in management research. Journal of Management 37(2): 610641.

66.

Nübold

Muck

Maier

(2013) A new substitute for leadership? Followers' state core self-evaluations. Leadership Quarterly 24(1): 2944.

67.

Piccolo

Colquitt

(2006) Transformational leadership and job behaviors: The mediating role of core job characteristics. Academy of Management Journal 49(2): 327340.

68.

Pillai

Schriesheim

Williams

(1999) Fairness perceptions and trust as mediators for transformational and transactional leadership: A two-sample study. Journal of Management 25(6): 897933.

69.

Podsakoff

MacKenzie

Moorman

(1990) Transformational leader behaviors and their effects on followers' trust in leader, satisfaction, and organizational citizenship behaviors. Leadership Quarterly 1(2): 107142.

70.

Podsakoff

MacKenzie

Lee

(2003) Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology 88(5): 879903.

71.

Purvanova

Bono

Dzieweczynski

(2006) Transformational leadership, job characteristics, and organizational citizenship performance. Human Performance 19(1): 122.

72.

Rauch

Frese

Sonnentag

(2000) Cultural differences in planning-success relationships: A comparison of small enterprises in Ireland, West Germany, and East Germany. Journal of Small Business Management 38(4): 2841.

73.

Shamir

House

Arthur

(1993) The motivational effects of charismatic leadership: A self-concept based theory. Organization Science 4(4): 577594.

74.

Shamir B, Zakay E, Breinin E and Popper M (1998) Correlates of charismatic leader behavior in military units: Subordinates' attitudes, unit characteristics, and superiors' appraisals of leader performance. Academy of Management Journal 41(4): 387–409.

75.

Siders

George

Dharwadkar

(2001) The relationship of internal and external commitment foci to objective job performance measures. Academy of Management Journal 44(3): 570579.

76.

Sitkin

Roth

(1993) Explaining the limited effectiveness of legalistic “remedies” for trust/distrust. Organization Science 4(3): 367392.

77.

Sluss

Ashforth

(2007) Relational identity and identification: Defining ourselves through work relationships. Academy of Management Review 32(1): 932.

78.

Taylor

MacKinnon

Tein

(2008) Tests of the three-path mediated effect. Organizational Research Methods 11(2): 241269.

79.

van Knippenberg

DeCremer

(2004) Leadership, self, and identity: A review and research agenda. Leadership Quarterly 15(6): 825856.

80.

Walumbwa

Avolio

Zhu

(2008) How transformational leadership weaves its influence on individual job performance: The role of identification and efficacy beliefs. Personnel Psychology 61(4): 793825.

81.

Walumbwa

Hartnell

(2011) Understanding transformational leadership-employee performance links: The role of relational identification and self-efficacy. Journal of Occupational and Organizational Psychology 84(1): 153172.

82.

Wofford

Whittington

Goodwin

(2001) Follower motive patterns as situational moderators for transformational leadership effectiveness. Journal of Managerial Issues 13(2): 196211.

83.

Yukl

(1989) Leadership in Organizations, Upper Saddle River, NJ: Prentice Hall.

84.

Zhu

Wang

Zheng

(2013) Examining the role of personal identification with the leader in leadership effectiveness: A partial nomological network. Group and Organization Management 38(1): 3667.