Are Formal Leaders the Only Ones Benefitting From Leadership Training? A Shared Leadership Perspective

Abstract

Leadership training most often involves training of formal leaders, and little is known about the potential benefits of leadership training for other members of an organization. Using theories of shared leadership, the current study examined outcomes of transformational leadership training that targets both formal and informal leaders (i.e., both vertical and shared leadership). The training was set in a Swedish paper pulp factory and involved formal and informal leaders participating in 20 days of training over a period of 16 months. Based on employee survey data collected both pre- and postintervention our analyses revealed that both formal and informal leaders significantly improved their transformational leadership behaviors. Interestingly, the improvement in transformational leadership behaviors of formal and informal leaders tended to predict employee efficiency and well-being in different ways. Improvements in formal leaders’ transformational leadership were related to employee well-being, while informal leaders’ increases in transformational leadership were associated with efficiency. The results point toward the benefit of a shared leadership perspective on leadership training and indicate that improvements in transformational leadership may affect employees differently depending on who in the organization displays them.

Keywords

shared leadership transformational leadership leadership training efficiency well-being

By tradition, leadership theory has been centered on the study of a singular, formal leader and focused on understanding how the behaviors or qualities of formal leaders affect followers and the organization (Bass, 2008). Recent approaches to the study of leadership have questioned this narrow focus on what has been labelled vertical leadership and argue that, as organizations have turned to team-based structures, it is important to understand leadership within teams as well (Nicolaides et al., 2014). Informal leadership provided by team members may also contribute to team outcomes, complementing the top-down influence of vertical leadership (D. Wang, Waldman, & Zhang, 2014). This approach, with leadership seen as originating not only from the designated leaders but also from team members themselves, has become known as shared leadership (Pearce, 2004).

Studies using theories of shared leadership have become increasingly popular during the past decade, and two recent meta-analyses suggest that shared leadership is positively related to team performance (Nicolaides et al., 2014) and team effectiveness (Wang et al., 2014). Also, meta-analytic evidence suggests that shared leadership explains unique variance over and above that of vertical leadership (Nicolaides et al., 2014). For example, in a study of new venture top management teams, leadership behaviors of team members positively predicted firm growth, controlling for vertical leadership (Ensley, Hmieleski, & Pearce, 2006). In a study of product development teams, shared leadership was more effective than purely vertical leadership for teams’ level of innovative behavior (Hoch, 2013). These findings may be explained by the fact that in many team contexts, vertical leaders may lack the human capital or the temporal resources needed to help their teams achieve their goals. In such contexts, shared leadership processes can provide support to vertical leadership efforts (Cohen & Bailey, 1997; Kerr & Jermier, 1978; Perry, Pearce, & Sims, 1999). Translating these findings into implications for leadership training, it could be argued that leadership training may be beneficial and important, not only for formal leaders but also for other team members (e.g., informal leaders). However, prior studies of leadership training have focused on the training of vertical leadership (e.g., Avolio, Reichard, Hannah, Walumbwa, & Chan, 2009), and less is known about whether leadership training for team members is worthwhile and how the training of team members affects followers in comparison to the training of vertical leaders.

The purpose of the present study was to examine employee-rated outcomes of transformational leadership training, which targeted both formal and informal leaders (i.e., both vertical and shared leadership). We make two important contributions to the literature on leadership training and shared leadership. First, we examine whether or not leadership training targeting both formal and informal leaders increases team members’ perception of their transformational leadership behaviors. This answers the call from Avolio et al. (2009) for further research on for whom and under what circumstances leadership training matters. Our study sheds light on whether leadership training of informal leaders can increase their leadership behavior toward their team members as much as leadership training of formal leaders can. Second, the study contributes to the literature by examining whether or not increased transformational leadership behaviors of informal leaders affects followers’ well-being and performance to the same extent as it does for formal leaders. This is important to understanding whether increased transformational leadership behavior of informal leaders actually affects team members. In addition, our study makes a general contribution to the leadership training literature by including follower outcomes, a naturalistic setting, and performance outcomes, all of which have been called for (Avolio et al., 2009).

Shared Leadership and the Importance of Informal Leadership

Interest in shared leadership and how team members’ leadership influence each other emerged as a consequence of the increased implementation of empowered teams and organizations with flatter structures (D. Wang et al., 2014). Shared leadership is based on the notion that leadership can be enacted by more than one member of a team (Morgeson, DeRue, & Karam, 2010). Conger and Pearce (2003) defined shared leadership as an ongoing, mutual influential process that involves peer, lateral, upward, or downward influences of team members. In comparison to vertical leadership, which involves the downward influence on subordinates by a formal leader (Conger & Pearce, 2003), shared leadership emphasizes social interaction among team members and is seen as a complement to traditional, singular forms of vertical leadership. Shared leadership, as an influence process, involves interacting with other members within the team and is manifested in behaviors such as communicating, influencing, making suggestions, and holding people accountable (Aime, Humphrey, DeRue, & Paul, 2014). These behaviors are an informal means of addressing the core leadership function typically addressed by formal leaders (Hoch & Kozlowski, 2014). Hence, shared leadership may be viewed as an emergent property of a team in which leadership roles and influence are distributed among team members (D’Innocenzo, Mathieu, & Kukenberger, 2016; Drescher, Korsgaard, Welpe, Picot, & Wigand, 2014)

The social identity theory of leadership (Hogg, 2001) has been used to explain why shared leadership is related to positive employee outcomes (e.g., performance). From a social identity theoretical standpoint, the most effective leader does not always have to be a formal leader; it can be other team members who better fit the actual in-group prototype of how a leader should act and behave (Hogg, 2001; Hogg, van Knippenberg, & Rast, 2012). The concept of shared leadership extends the prototypicality of leadership when team members are able to take on a leadership role and, over time, more leader behavior in line with the in-group prototype should be associated with members acting in a manner that will be best for the group in both the short and long run (Hogg, 2001; Hogg et al., 2012).

As with vertical leadership, shared leadership can take on a number of different forms. Consistent with the notion that shared leadership includes the distribution of leadership functions among team members, a number of studies have focused on shared leadership in relation to different leadership functions, such as information search and structuring, information use in problem solving, and managing personnel and material resources (e.g., Drescher et al., 2014). Another common approach to conceptualizing shared leadership is to assess the extent to which traditional leadership behaviors are performed by team members. For example, Pearce (2004) differentiated between four basic types of shared leadership that can be demonstrated by the members of a team: directional, empowering, transactional, and transformational. Using Pearce’s model, a team member could demonstrate shared directive leadership by assigning responsibilities to other team members or clarifying exchanges needed to gain group decisions. Shared empowering leadership could be displayed by the peer-based encouragement of self-leadership in others. Shared transactional leadership involves one or more team members recognizing and rewarding the efforts, contributions, or collaboration of others in the team. In a team, this could be done by collegial appraisal for achievements. Another way to reward people is to give colleagues the opportunity to recommend individuals or teams for different kinds of recognition when they reach different targets or goals (Gockel & Werth, 2011; Pearce, 2004). Shared transactional leadership, with focus on recognition and reward, is therefore a broader construct than vertical transactional leadership described by Bass (see, e.g., Bass & Riggio, 2006), where focus has been on transactions and exchanges and on material rewards.

Shared transformational leadership, which is the focus of this study, can be demonstrated by peer-based inspiration or the articulation of a unifying vision. Recent studies have demonstrated that vision formation is indeed more often a collective process involving multiple individuals, rather than the work of a formal leader (Ensley et al., 2006; Pearce & Ensley, 2004). Team members may also display other transformational behaviors, such as intellectual stimulation in terms of questioning assumptions and individual consideration by listening attentively to their colleagues’ problems. While transformational leaders (i.e., vertical leadership) inspire and challenge team members to build their confidence, shared transformational leadership involves team members having influence over each other. Consequently, shared leadership is about accepting one another as leaders as opposed to being, for example, self-empowered by a single leader. Thus, a formal leader may be transformational but retain most of the authority and influence (Carson, Tesluk, & Marrone, 2007). In line with empirical studies on the effectiveness of vertical transformational leadership, studies of shared transformational leadership suggest that team members’ display of transformational leadership behaviors are positively related to team effectiveness (Wang et al., 2014), new venture growth (Ensley et al., 2006), trust, and team potency (Boies, Lvina, & Martens, 2010). Given that research so far has provided evidence that shared transformational leadership can have a potential effect on group behavior, attitudes, cognition, and performance (Pearce, Conger, & Locke, 2008), the next important step would be to investigate whether shared transformational leadership can be trained.

Transformational Leadership Training of Formal and Informal Leaders

Leadership training has been suggested to be an effective way to develop leadership skills and improve organizations. A recent meta-analysis suggests that the effort and financial resources that are invested in leadership training are worthwhile. Avolio et al. (2009) found that leaders who participate in leadership training have a higher probability of improved leadership compared to leaders who do not participate in training, demonstrating that leadership training in general is useful. In relation to transformational leadership theory, a number of intervention studies have focused on the effects of transformational leadership training, indicating that transformational leadership behavior is both trainable and can positively affect employee outcomes. To our knowledge, five field-based interventions have examined the effectiveness of transformational leadership training: Barling, Weber, and Kelloway (1996), Hardy et al. (2010), Kelloway, Barling, and Helleur (2000), Dvir, Eden, Avolio, and Shamir (2002), and Popper, Landau, and Gluskinos (1992). For example, Barling et al. (1996) conducted a randomized field experiment including 20 managers in the Canadian financial sector. The training included 1 day of training followed by four individual booster sessions where managers received feedback on their transformational leadership behavior and developed action plans for improving their leadership. The analyses demonstrated that the transformational leadership training positively enhanced followers’ perception of their leaders’ transformational behaviors, as well as follower commitment and performance. In a similar study, Dvir et al. (2002) demonstrated that transformational leadership training in the Israeli army positively affected followers’ perception of their leaders’ behaviors and increased follower self-efficacy, critical independent approach, collectivistic orientation, and work performance.

To sum up, prior intervention studies suggest that transformational leadership training positively affects followers’ perception of their leaders’ transformational leadership behaviors and followers’ attitudes and performance. Given that transformational leadership from team members has been proven to affect employees over and above that of the formal leader (Nicolaides et al., 2014), it is of interest to investigate whether transformational leadership training of others besides formal leaders, such as individual team members, would be equally effective. To the best of our knowledge, no studies have determined whether transformational leadership training of informal leaders or team members may be an important complement to the training of formal leaders.

The Present Study

The aim of the present study was to compare the effects of an 18-month long leadership training program in the forest industry with a focus on transformational leadership between a group of formal leaders and a group of informal leaders who worked as technical engineers embedded in the work teams. The technical engineers had no formal leadership role, but they were responsible for machines and production and worked embedded in the work teams.

The present study seeks to contribute both to the leadership training literature in general and the understanding of the shared transformational leadership training in specific. Based on our review of shared leadership theory and previous research on leadership training, we propose that the leadership training will improve transformational leadership behaviors in terms of idealized influence, inspirational motivation, intellectual stimulation, and individualized consideration for both formal and informal leaders who participate in the training.

Hypothesis 1a: Transformational leadership training will increase followers’ perception of the transformational leadership of formal leaders.

Hypothesis 1b: Transformational leadership training will increase followers’ perception of the transformational leadership of informal leaders.

Based on transformational leadership theory, we also expect that increases in transformational leadership will affect more distal outcomes related to employee well-being and performance (Bass & Riggio, 2006). Transformational leadership behaviors may influence employee well-being in a number of ways, for example, if leaders reframe stressful situations into developmental opportunities (intellectual stimulation), replace feelings of isolation with a sense of belonging (idealized influence), or listen carefully to the personal needs of their followers (individualized influence; Bass & Riggio, 2006). Empirical studies have supported a positive relationship between transformational leadership and employee well-being (e.g., DeGroot, Kiker, & Cross, 2000; Harms, Credé, Tynan, Leon, & Jeung, 2017), including job satisfaction (Braun, Peus, Weisweiler, & Frey, 2013) and negative relationship between transformational leadership and sick leave (Nielsen & Daniels, 2016). We therefore propose:

Hypothesis 2a: Increases in transformational leadership after training will be positively related to job satisfaction among the followers of formal leaders.

Hypothesis 2b: Increases in transformational leadership after training will be positively related to job satisfaction among the followers of informal leaders.

Hypothesis 3a: Increases in transformational leadership after training will be negatively related to sick leave among the followers of formal leaders.

Hypothesis 3b: Increases in transformational leadership after training will be negatively related to sick leave among the followers of informal leaders.

Transformational leadership behaviors are also expected to increase employee performance. By communicating a compelling vision of the future (inspirational motivation), followers view their work as more meaningful and significant and thus increase their intrinsic motivation. Transformational leaders instill a belief that goals can be achieved while providing the support and coaching that followers need to accomplish their jobs (Bass & Riggio, 2006). Meta-analyses have confirmed a positive relationship between transformational leadership and employee performance (e.g., Lowe, Kroeck, & Sivasubramaniam, 1996; G. Wang, Oh, Courtright, & Colbert, 2011). We therefore propose:

Hypothesis 4a: Increases in transformational leadership after training will be positively related to efficiency among the followers of formal leaders.

Hypothesis 4b: Increases in transformational leadership after training will be positively related to efficiency among the followers of informal leaders.

Method

Participants and Procedure

The study was set in a Swedish forest industry company with approximately 800 employees. All of the company’s formal leaders participated in the leadership training. Formal leaders were defined as the management level that is directly above nonmanagerial workers with responsibility for employee and work environment issues. The company also decided to send all technical engineers to training, who in this study constituted our group of informal leaders. Technical engineers worked embedded in the teams and were responsible for machines and production. Although they had no formal leadership responsibility, the company saw the group of technical engineers as key team members who, given their knowledge and skills, might find themselves in situations where they could take on a leadership role. They could thereby face a difficult balancing act with the possibility of influencing team members but without a formal mandate, hence leadership training might be helpful.

Out of the 101 formal and informal leaders, 74 completed both the pre- and postintervention questionnaires, which were administered 1 month before and after training, respectively, and gave informed consent for the data to be used in research (response rate 73%). Of these leaders, 47 were formal and 27 were informal. Among the formal leaders, 76% were men, and the mean age was 47 years (SD = 8.2); they had an average of 6 years of working in their current position (SD = 6.3) and 20 years (SD = 11.1) in the organization. In the group of informal leaders, 76% were men, the mean age was 45 years (SD = 9.7), and they had worked for 6 years in their current position (SD = 5.0) and 18 years (SD = 12.2) in the organization. An independent samples t test revealed no significant differences between the group of formal and informal leaders on these demographic variables.

In the second step of recruitment, the leaders were asked to invite five of their team members to complete the evaluation questionnaire. They were instructed to invite both team members to whom they felt close and those whom they perceived as more distant. This procedure was used since the intervention included multisource (360-degree) feedback on the leaders’ leadership. A total of 290 team members were invited, 237 of whom answered the questionnaire (response rate of 90%) and gave informed consent for use of the data in research. The final employee sample consisted of 211 individuals after those who did not provide sufficient data on the questionnaire items were eliminated. In the final sample, 82% were men, the mean age was 46 years (SD = 8.9), and they had worked within the organization for a mean of 23 years (9.9 years). Given that the informal leaders had no employees of their own but rather were embedded in the work teams, some employees were, by necessity, invited by both their formal and informal leaders; in the end, 56 of the 211 employees rated both their formal and informal leaders.

Measures

Transformational Leadership

We measured transformational leadership as rated by employees with the Multifactor Leadership Questionnaire (MLQ-Form 5X; Avolio & Bass, 2004). Twenty items represented the four subcomponents of transformational leadership: idealized influence, inspirational motivation, intellectual stimulation, and individualized consideration. A 5-point response scale ranging from 1 (never) to 5 (frequently, if not always) was used. An example of an item is “My supervisor articulates a compelling vision of the future.” The internal consistency of the MLQ subscales at baseline ranged between .70 and .83.

Efficiency

Self-rated efficiency was measured with the work efficiency subscale from the Health and Work Questionnaire (Shikiar, Halpern, Rentz, & Khan, 2004). For each subscale, the respondent was asked to make ratings from three different perspectives. Efficiency was measured with the item “How would you and the following persons describe your work efficiency during the last week?” and there were separate response alternatives for ratings from the perspective of the respondent and of his/her supervisor and coworkers. The response alternatives were measured on a 10-point Likert-type scale with from my worst ever and my best ever as endpoints, and the scale reliability was .88.

Well-Being

We used two indicators of employee well-being: job satisfaction and self-rated sick leave. Job satisfaction was evaluated with a single item, “I am satisfied with my job” (Hellgren, Sjöberg, & Sverke, 1997). A 4-point response scale ranging from 1 (does not agree) to 5 (fully agree) was used. Sick leave was measured by employees’ rating of the total number of days absent from work due to their own illness during the past 6 months (Allvin et al., 1999). Five categories were presented: not absent due to illness, 1 to 7 days, 8 to 30 days, 31 to 90 days, and more than 90 days.

The Leadership Training

The leadership training ran over a period of 16 months (December 2011 to March 2013). Each participant received 20 days of training in cross-departmental groups of about 20 leaders. Following recommendations for effective leadership training (Cacioppe, 1998), multiple training methodologies were employed. The training was based on transformational leadership theory combined with applied behavioral analysis (ABA; Skinner, 1963) to equip managers with knowledge about how they may influence employee behavior.

The training began with individual feedback on the 360-degree evaluation of the participating leaders’ leadership behaviors and action planning based on this feedback assessment. Thereafter, the training consisted of two blocks of training: one theoretical and one practical. The theoretical block included 12 full days and 4 half days of training from December 2011 to June 2012. It was organized into six 2-day sessions with different themes, with half-day follow-up practical sessions conducted between them. The first session included exercises to increase participants’ motivation to change, lectures on transformational leadership and ABA, and exercises and discussions about their ideal leader in relation to a transformational leader. The following half day consisted of discussions about the 360-degree assessment for which they worked in pairs to coach each other. The second session focused on behavior and included lectures, exercises, and a group discussion on how ABA could help them understand leadership and employee behavior. The following half day included group exercises and discussions about changing behaviors. The third session focused on breaking down the company’s overall vision into the different transformational leadership behaviors that the leaders could display in their daily work. It also included exercises on communication and on how to give feedback. The fourth session included connecting transformational leadership to productivity and safety, two important goals for the company. It included both lectures and exercises. During the following half day, the format of the action plans was introduced, and during the fifth session and the following half day the leaders worked on their individual action plans. The last session, Session 6, was devoted to summarizing the program and what they had learned as well as celebrating successes.

The practical block ran between August 2012 and March 2013 and included 6 full days of training. The leaders each identified an area for improvement that they wanted to work through on site, together with their work groups, while applying what they had learned from the theoretical block. Examples of improvement areas that leaders chose included transformational leadership, collaboration and information-sharing in the team, and safety management. The leaders received feedback and support on their improvement projects, which they worked on between sessions. In addition, the practical block included the establishment of internal follow-up routines in the organization in an effort to make sure that the improvement projects were followed up with and coordinated internally. The leadership training was developed and carried out by the participating organization’s local occupational health service. The researchers designed the outcome evaluation and the data collection.

Analysis

The employee data collected before and after the training were first analyzed using separate, two-factor Group × Time analyses of variance with repeated measures on the second factor. Given that multiple ANOVAs increase the chances of making a Type 1 error by capitalizing on chance, it may be sensible to control for this by either running preliminary MANOVAs or by using Bonferroni to correct the significance level that is used. However, in line with Hardy et al. (2010), we took neither of these courses of action for three reasons: (a) There was a risk that our main and interaction effects could be “washed out” in a MANOVA by the nonsignificant variables in each effect. (b) Clear directional a priori hypotheses had been specified for our variables. If our patterns of effects were repeatedly obtained for our hypothesized variables, then they clearly could not have been obtained by chance, as chance effects would be random in direction. (c) There were no hypotheses about the combined linear effects of transformational leadership behaviors. It is only appropriate to use MANOVA if genuinely multidimensional hypotheses have been formulated.

Prior to our multiple group latent difference score (LDS) analysis, a longitudinal multiple-group CFA model was estimated with the four subscales included as indicators of the two latent variables—employees’ perception of transformational leadership before and after the intervention. In line with current recommendations (Little, 2013; Widaman, Ferrer, & Conger, 2010), configural, metric (factor loadings), and scalar (factor loadings and intercepts) invariance was tested across group and time in the measurement model. We used change in the comparative fit index (ΔCFI) as the goodness of fit index when comparing the models. A decrease equal to or greater than −.01 in CFI is considered as an indication of noninvariance (Cheung & Rensvold, 2002). If metric, but not scalar, invariance is reached, partial measurement invariance can be established (Steenkamp & Baumgartner, 1998). As long as there are at least two loadings and intercepts that are constrained as equal across groups and time, we can make valid inferences about the differences between latent factor means in the model (Byrne, Shavelson, & Muthén, 1989).

We then specified a multiple group LDS model (e.g., McArdle, 2009; McArdle & Nesselroade, 1994; Selig & Preacher, 2009; Wu, Selig, & Little, 2013). The LDS model explicitly estimates interindividual differences in true intraindividual change in the targeted variables (Selig & Preacher, 2009; Wu et al., 2013). In our two-wave LDS model, a latent difference variable represents the difference between transformational leadership at Time 1 and transformational leadership at Time 2 corrected for measurement error. The latent difference variable is specified by regressing the variable score at Time 2 on the same variable at Time 1 and fixing the unstandardized path to 1. In addition, the variable score at Time 2 is regressed on the latent difference variable and fixing the unstandardized path to 1. The initial level at Time 1 and the latent difference variable are allowed to covary. The multiple group latent changes score analyses were carried out with Mplus Version 7.31 (Muthén & Muthén, 2012) with maximum likelihood estimation.

To evaluate the model fit, we used conventional fit indices including the CFI, the Tucker–Lewis Index (TLI), the standardized root mean residual (SRMR), and the root mean square error of approximation (RMSEA). Traditional cutoff criteria (CFI and TLI >.90, SRMR and RMSEA < .08) were used to indicate acceptable fit (Kline, 2010; Marsh, 2007), and Hu and Bentler’s (1999) more restrictive criteria (CFI and TLI > .95, SRMR and RMSEA < .06) were used to indicate very good fit.

Results

The descriptive results for the two intervention groups at baseline and follow-up based on employee ratings are displayed in Table 1. There was a trend for scores of transformational leadership to increase across the period of the study for both groups. Furthermore, the mean scores across the two groups tended to be similar.

Table 1.

Descriptives and F Values for the 2 × 2 Mixed Model ANOVAs (Group × Time) of Leadership Behaviors.

	Formal leaders		Informal leaders		F for Group	F for Time	F for Interaction
	Pretest, M (SD)	Posttest, M (SD)	Pretest, M (SD)	Posttest, M (SD)	F for Group	F for Time	F for Interaction
Leadership behavior
Idealized influence	2.29 (0.70)	2.45 (0.71)	2.24 (0.58)	2.41 (0.66)	0.30	23.06***	0.00
Inspirational motivation	2.33 (0.81)	2.50 (0.80)	2.27 (0.66)	2.37 (0.72)	1.23	10.91**	0.68
Intellectual stimulation	2.25 (0.68)	2.57 (0.69)	2.39 (0.58)	2.63 (0.59)	2.00	52.22***	1.02
Individualized consideration	2.12 (0.84)	2.46 (0.71)	2.24 (0.70)	2.53 (0.64)	1.53	57.33***	0.47

Note. ANOVAs = analyses of variance. Formal leaders n = 47 (157 employee ratings); informal leaders n = 27 (110 employee ratings).

p < .05. **p < .001. ***p < .0001.

The Group × Time ANOVAs indicated that the Time effects mentioned above were significant for all four transformational leadership behavior scales (see Table 1). The largest effects were obtained for individualized consideration, F(1, 267) = 57.33, p < .001, and intellectual stimulation, F(1, 267) = 52.22, p < .001, but idealized influence and inspirational motivation also increased significantly. Also, notably, there were no significant main effects for Group nor any interaction effects of Group and Time on any of the transformational leadership behaviors. This suggests that there were no differences in mean scores between the two groups and that the positive effects of the leadership training measured from the perspective of employees did not differ between the formal and informal leaders, supporting Hypotheses 1a and 1b.

Next, a multiple group latent change score model that compared whether or not change in leadership behaviors affected employees differently if the trainee was a formal or informal leader was estimated. Prior to estimating the models, longitudinal measurement invariance was examined to make sure that our measures were stable over time and across groups (Little, 2013; Widaman et al., 2010). As presented in Table 2, increasingly restricted models were estimated, the first without any equality constraints (configural invariance), the second with equality constraints on the factor loadings (metric invariance), and the third with equality constraints on factor loadings and intercepts (scalar invariance; Little, 2013). Given that metric, but not scalar, invariance was supported, partial measurement invariance was examined with two items with invariant loadings and intercepts on each factor. To establish partial measurement invariance, the intercepts of intellectual stimulation and individualized consideration over time were released.

Table 2.

Longitudinal Measurement Invariance Across Groups (N = 267).

Model	χ²	df	CFI	RMSEA	90% CI
Model	χ²	df	CFI	RMSEA	Lower	Upper
Configural	42.827	26	.989	.069	.028	.105
Metric	53.050	35	.988	.062	.022	.094
Scalar	105.307	44	.959	.102	.077	.127
Partial	69.356	42	.982	.070	.038	.098

Note. df = degrees of freedom; CFI = comparative fit index; RMSEA = root mean square error of approximation; CI = confidence interval.

Following the establishment of measurement invariance, the latent change score model was estimated in both groups, now including three employee-rated outcomes: job satisfaction, sick leave, and efficiency measured at Time 2. The estimated model demonstrated an acceptable fit: χ² (85) = 145.54, p < .001, CFI = .96, TLI = .95, SRMR = .10, RMSEA = .07, 90% confidence interval [CI] [.05, .09]. As presented in Figure 1, change in transformational leadership among formal leaders was positively associated with Time 2 employee job satisfaction (β = .27, p < .05), supporting Hypothesis 2a, but unrelated to employee self-rated sick leave and efficiency, contradicting Hypotheses 3a and 4a. For informal leaders, change in transformational leadership was positively related to Time 2 employee efficiency (β = .23, p < .05), supporting Hypothesis 4b, but unrelated to employee job satisfaction or self-rated sick leave, thereby contradicting Hypotheses 2b and 3b. To assess whether the differences in parameters between the two groups were significant, Wald’s test of parameter constraints was employed to compare the estimates of job satisfaction, self-rated sick leave, and efficiency. The results revealed that although the parameters in the groups differed, these differences were not significant.

Figure 1.

Multiple group latent change score model of change in transformational leadership after leadership training and associations with employee outcomes.

Discussion

This study set out to examine for whom and under what circumstances leadership training matters by comparing the training outcomes of formal and informal leaders sent to training. Our analysis shows improvements in employees’ ratings for both formal and informal leaders’ transformational leadership over time as the leaders participated in leadership training. There were no differences between how the formal and informal leaders developed in transformational leadership behaviors in any of the transformational leadership subscales from the employees’ perspective. The hypothesis that changes in transformational leadership would be related to improvement in employee distal outcomes was partially supported. Among formal leaders, increase in transformational leadership behaviors was related to improved employee job satisfaction but not efficiency, and vice versa for informal leaders. Changes in transformational leadership behaviors were not related to sickness absence in either group. In the following section, we discuss the findings relating each of the hypotheses and then the practical implications of the findings.

In line with the shared leadership theory (Conger & Pearce, 2003; Pearce, 2004), we found support for our first hypothesis: Transformational leadership training increased followers’ perception of transformational leadership of both formal and informal leaders. The empirical support for the shared leadership theory is primarily based on correlational studies, however (e.g., Wang et al., 2014). Our intervention study confirms those findings and extends previous research by demonstrating that transformational leadership can be experienced in others besides formal leaders.

Also, both formal and informal leaders can develop their transformational leadership behaviors, such that changes in transformational leadership among formal and informal leaders alike can be detected by employees. The fact that the transformational leadership measures demonstrated strong measurement invariance across groups (but only partial measurement invariance over time) supports this, thus implying that employees rated the transformational leadership behavior of both formal and informal leaders in a similar way. It also suggests that employees may perceive and detect leadership behaviors in others besides formal leaders in a reliable matter. This indicates that training in transformational leadership is related to improvements in transformational leadership that are as strong among informal leaders as formal ones.

Previous research has shown that transformational leadership training is positively related to employee outcomes (e.g., Barling et al., 1996; Hardy et al., 2010; Kelloway et al., 2000). The results from this study add to this knowledge by lending some, but not unequivocal, support to the relationship between transformational leadership training and employee outcomes. Change in transformational leadership was unrelated to employee self-rated sick leave in both groups, contradicting Hypothesis 3. For Hypotheses 2 and 4, which stated that changes in transformational leadership were expected to be related to improvements in employee job satisfaction and efficacy, the results seem to indicate a differential pattern for formal and informal managers. Whereas an increase in formal leaders’ transformational leadership was related to employee job satisfaction (supporting Hypothesis 2a and contradicting Hypothesis 4a), an increase in informal leaders’ transformational leadership was related to employee efficiency (contradicting Hypothesis 2b and supporting Hypothesis 4b).

Yet the parameters were not statistically significantly different. Therefore, more research is needed to determine whether the difference in pattern represents a true difference in how transformational leadership training relates to employee outcomes among formal and informal leaders. Nevertheless, the pattern of results does have some face validity given the leaders’ respective responsibilities and tasks. The formal leaders have personnel responsibilities. In Sweden, where this study was set, this includes formal responsibility for the physical and psychosocial work environment, such as striving for employee well-being in a broad sense, of which job satisfaction is an indicator. The informal leaders in this specific setting had responsibilities for daily operations that were closely related to efficiency. Therefore, one interpretation of the results is that transformational leadership may be directed toward influencing certain outcomes, depending on the targets of the leader. This ties in to the current discourse on general versus specific transformational leadership, which shows that for outcomes such as safety, transformational-specific leadership may be more important than general transformational leadership (Barling, Loughlin, & Kelloway, 2002). Although this study was not set up to test general versus specific leadership, it does indicate that, for leaders with different responsibilities, transformational leadership may be related to different types of employee outcomes. Overall, the results indicate that some of the improvements in employee distal outcomes following from increased transformational leadership hold up for informal leaders in addition to formal ones. Future studies are needed to illuminate this further and to examine whether other groups of employees may also benefit from leadership training.

Practical Implications

The present study offers a number of practical implications for organizations to consider in relation to leadership training. In line with transformational leadership theory (Bass & Riggio, 2006) and the few previous studies of transformational leadership training (e.g., Hardy et al., 2010; Barling et al., 1996), our study demonstrates that transformational leadership can be trained and that improvement in transformational leadership is related to positive employee outcomes. Our findings suggest that transformational leadership training is an efficient intervention for improving important employee outcomes that may offer an alternative to large-scale organizational interventions to improve, for example, employee health and well-being. In this respect, our study answers the call from Kelloway and Barling (2010) to examine leadership training as a potential occupational health intervention.

Furthermore, in line with shared leadership theory (Conger & Pearce, 2003; Pearce, 2004), our study suggests that it may be fruitful for organizations to send not only formal leaders to training but also others, such as those with responsibility for production. By tradition, most organizations send only formal leaders to training, thus putting a heavy burden on their shoulders to take on the sole leadership responsibility for their teams. Building on shared leadership theory (Conger & Pearce, 2003; Pearce, 2004) and the results from our study, we encourage organizations to reconsider this narrow, hierarchical view of leadership and instead expand their view to see the full leadership potential in their organizations by allowing leadership responsibilities to be shared among group members. By carefully analyzing which employees or groups of employees may have the highest potential to influence their fellow coworkers, organizations can identify additional leadership potential in employees who may also benefit from leadership training. In addition, opening up the possibility for other employees besides the formal leader to attend leadership training sends a message to workers that others are also welcome to take on leadership roles and responsibilities in the organization and that everyone is expected to help lead the organization in its envisioned direction. Although our leadership training focused on individual leaders and their development, our findings suggest that others besides the formal leader also contribute to the leadership process. This indicates that it may be worthwhile for organizations to broaden their view of leadership development beyond the training of individuals by focusing on the whole leadership process, as suggested by Day, Fleenor, Atwater, Sturm, and McKee (2014).

Limitations

There are some important limitations that merit consideration in appraising the findings of the present study. First, there no comparison group that did not participate in the leadership training. The design with data collected both pre- and postintervention with two active groups (formal and informal leaders) does allow for the detection of changes over time, as well as between the two groups, but without randomization and a control group, causal inferences cannot be made (Shannon, Robson, & Guastello, 1999). Establishing an appropriate control group, in addition to using randomization, is a common challenge in organizational intervention research where the design of the study is a result of a negotiation with the participating organization and limited by practical constraints. In this case, all formal and informal leaders participated in the study, limiting the possibility for establishing a control group.

We took the following steps to offset the disadvantages with the noncontrolled design. First, we mitigated the influence that participation in itself may have had on outcomes by using ratings from persons who did not participate in the intervention (the leaders participated and employees rated them). Second, we used a priori hypotheses relating a certain change (i.e., in transformational leadership behavior) to employee outcomes. The fact that change in transformational leadership was significantly related to changes in some employee outcomes offers some, but not conclusive, evidence that changes in leadership behavior are related to changes in employee behavior. Yet the difficulty in establishing with certainty why that change occurred persists. As mentioned above, this study included all formal and informal leaders in the organization. This, in combination with a high participation rate among leaders and leaders who agreed to participate in research, means that the conclusions are based on a quite unique sample representing a full range of leaders in an organization, not only those who volunteered to participate in a leadership training. This means that if anything, the impact of the leadership training is underestimated. Another limitation is the use of a single item to measure job satisfaction, as the reliability cannot be established. Yet single items of job satisfaction have proven to have sound psychometric properties (Fisher, Matthews, & Gibbons, 2016), offsetting some of the drawbacks of using a single item.

Conclusions

Even with these noted limitations, our study makes two important contributions to the literature. First, by using a shared leadership perspective on leadership training, we demonstrate that other personnel besides formal leaders may benefit from leadership training, as our group of informal leaders working as engineers also increased their transformational leadership behavior during the training they received. Second, the present study opens up the possibility that improvements in transformational leadership may affect employees differently, depending on who in the organization displays them. This calls for a new avenue of research examining how leaders and employees in different positions may use transformational leadership behaviors to enhance the efficiency and well-being of employees.

Future research may examine if improvement in other types of shared leadership behaviors, including directional, empowering, and transactional leadership, also affect employees differently depending on who displays them. Finally, given the focus on teams in theories of shared leadership, it may be fruitful to also explore other types of training methods to increase shared leadership, such as team training.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a postdoc grant to Susanne Tafvelin from FORTE (Grant No. 2014-0739) and a personal grant to Ulrica von Thiele Schwarz through a fellowship in Improvement Science from Vinnvård (Grant No. VF13-008).

Author Biographies

Susanne Tafvelin is an associate professor at the Department of Psychology at Umeå University, Sweden. Her present research interests include transfer of leadership training, the implementation of occupational health interventions in organizations and the relationship between leadership and employee health.

Henna Hasson is an associate professor at Karolinska Institutet, and head of the Unit for Implementation at the Centrum for Epidemiology and Community Medicine at the Stockholm County Council, Sweden. Her research focus on the implementation of occupational health interventions.

Stefan Holmström is a senior lecturer at the Department of Psychology at Umeå University. His present research interests include well-being and performance both at work and within sports.

Ulrica von Thiele Schwarz is a professor at Mälardalen University and associate professor at Karolinska Institutet, Sweden. Her research focuses on improvement in organizations with a special interest in organizational level interventions to improve employee health.

References

Aime

Humphrey

DeRue

D. S.

Paul

J. B.

(2014). The riddle of heterarchy: Power transitions in cross-functional teams. Academy of Management Journal, 57, 327-352.

Allvin

Aronsson

Hagström

Hansson

Johansson

Lundberg

Skärstrand

(1999). Frågeformuläret STEM. Gränslöst arbete eller arbetets nya gränser [The STEM questionnaire. Boundaryless work or the new boundaries of work]. Stockholm, Sweden: Arbetslivsinstitutet [National Institute for Working Life].

Avolio

B. J.

Bass

B. M.

(2004). MLQ: Multifactor leadership questionnaire. Menlo Park, CA: Mind Garden.

Avolio

B. J.

Reichard

R. J.

Hannah

S. T.

Walumbwa

F. O.

Chan

(2009). A meta-analytic review of leadership impact research: Experimental and quasi-experimental studies. Leadership Quarterly, 20, 764-784.

Barling

Loughlin

Kelloway

E. K.

(2002). Development and test of a model linking safety-specific transformational leadership and occupational safety. Journal of Applied Psychology, 87, 488-496.

Barling

Weber

Kelloway

E. K.

(1996). Effects of transformational leadership training on attitudinal and financial outcomes: A field experiment. Journal of Applied Psychology, 81, 827-832.

Bass

B. M.

(2008). The Bass handbook of leadership: Theory, research and managerial applications (4th ed.). New York, NY: Simon & Schuster.

Bass

B. M.

Riggio

R. E.

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

Boies

Lvina

Martens

M. L.

(2010). Shared leadership and team performance in a business strategy simulation. Journal of Personnel Psychology, 9, 195-202.

10.

Braun

Peus

Weisweiler

Frey

(2013). Transformational leadership, job satisfaction, and team performance: A multilevel mediation model of trust. Leadership Quarterly, 24, 270-283.

11.

Byrne

B. M.

Shavelson

R. J.

Muthén

(1989). Testing for the equivalence of factor covariance and mean structures: The issue of partial measurement invariance. Psychological Bulletin, 105, 456-466.

12.

Cacioppe

(1998) Leaders developing leaders: An effective way to enhance leadership development programs. Leadership & Organization Development Journal, 19, 194-198.

13.

Carson

J. B.

Tesluk

P. E.

Marrone

J. A.

(2007). Shared leadership in teams: An investigation of antecedent conditions and performance. Academy of Management Journal, 50, 1217-1234.

14.

Cheung

G. W.

Rensvold

R. B.

(2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling, 9, 233-255.

15.

Cohen

Bailey

D. E.

(1997). What makes teams work: Group effectiveness research form the shop floor to the executive suite. Journal of Management, 23, 239-290.

16.

Conger

J. A.

Pearce

C. L.

(2003). Shared leadership: Reframing the hows and whys of leadership. Thousand Oaks, CA: Sage.

17.

Day

D. V.

Fleenor

J. W.

Atwater

L. E.

Sturm

R. E.

McKee

R. A.

(2014). Advances in leader and leadership development: A review of 25years of research and theory. Leadership Quarterly, 25, 63-82.

18.

D’Innocenzo

Mathieu

J. E.

Kukenberger

M. R.

(2016). A meta-analysis of different forms of shared leadership–team performance relations. Journal of Management, 42, 1964-1991.

19.

DeGroot

Kiker

D. S.

Cross

T. C.

(2000). A meta-analysis to review organizational outcomes related to charismatic leadership. Canadian Journal of Administrative Sciences, 17, 356-371.

20.

Drescher

M. A.

Korsgaard

M. A.

Welpe

I. M.

Picot

Wigand

R. T.

(2014). The dynamics of shared leadership: Building trust and enhancing performance. Journal of Applied Psychology, 99, 771-783.

21.

Dvir

Eden

Avolio

B. J.

Shamir

(2002). Impact of transformational leadership on follower development and performance: A field experiment. Academy of Management Journal, 45, 735-744.

22.

Ensley

M. D.

Hmieleski

K. M.

Pearce

C. L.

(2006). The importance of vertical and shared leadership within new venture top management teams: Implications for the performance of startups. Leadership Quarterly, 17, 217-231.

23.

Fisher

G. G.

Matthews

R. A.

Gibbons

A. M.

(2016). Developing and investigating the use of single-item measures in organizational research. Journal of Occupational Health Psychology, 21, 3-23.

24.

Gockel

Werth

(2011). Measuring and modeling shared leadership. Journal of Personnel Psychology, 9, 172-180.

25.

Hardy

Arthur

C. A.

Jones

Shariff

Munnoch

Isaacs

Allsopp

A. J.

(2010). The relationship between transformational leadership behaviors, psychological, and training outcomes in elite military recruits. Leadership Quarterly, 21, 20-32.

26.

Harms

P. D.

Credé

Tynan

Leon

Jeung

(2017). Leadership and stress: A meta-analytic review. Leadership Quarterly, 28, 178-194.

27.

Hellgren

Sjöberg

Sverke

(1997). Intention to quit: Effects of job satisfaction and job perceptions. In Avallone

Arnold

de Witte

(Eds.), Feelings work in Europe (pp. 415-423). Milano, Italy: Guerini.

28.

Hoch

J. E.

(2013). Shared leadership and innovation: The role of vertical leadership and employee integrity. Journal of Business and Psychology, 28, 159-174.

29.

Hoch

J. E.

Kozlowski

S. W.

(2014). Leading virtual teams: Hierarchical leadership, structural supports, and shared team leadership. Journal of Applied Psychology, 99, 390-403.

30.

Hogg

M. A.

(2001). A social identity theory of leadership. Personality and Social Psychology Review, 5, 184-200.

31.

Hogg

M. A.

van Knippenberg

Rast

D. E.

III . (2012). The social identity theory of leadership: Theoretical origins, research findings, and conceptual developments. European Review of Social Psychology, 23, 258-304.

32.

L. T.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.

33.

Kelloway

E. K.

Barling

(2010). Leadership development as an intervention in occupational health psychology. Work & Stress, 24, 260-279.

34.

Kelloway

K. E.

Barling

Helleur

(2000). Enhancing transformational leadership: The roles of training and feedback. Leadership & Organization Development Journal, 21, 145-149.

35.

Kerr

Jermier

J. M.

(1978). Substitutes for leadership: Their meaning and measurement. Organizational Behavior and Human Performance, 22, 375-403.

36.

Kline

R. B.

(2010). Principles and practice of structural equation modeling (3rd ed.). New York, NY: Guilford Press.

37.

Little

T. D.

(2013). Longitudinal structural equation modeling. New York, NY: Guilford Press.

38.

Lowe

K. B.

Kroeck

K. G.

Sivasubramaniam

(1996). Effectiveness correlates of transformational and transactional leadership: A meta-analytic review of the MLQ literature. Leadership Quarterly, 7, 385-425.

39.

Marsh

H. W.

(2007). Application of confirmatory factor analysis and structural equation modeling in sport/exercise psychology. In Tenenbaum

Eklund

R. C.

(Eds.), Handbook of sport psychology (3rd ed., pp. 774-798). Hoboken, NJ: Wiley.

40.

McArdle

J. J.

(2009). Latent variable modeling of differences and changes with longitudinal data. Annual Review of Psychology, 60, 577-605.

41.

McArdle

J. J.

Nesselroade

J. R.

(1994). Using multivariate data to structure developmental change. In Cohen

S. H.

Reese

H. W.

(Eds.), Life-span developmental psychology: Methodological contributions (pp. 223-267). Hillsdale, NJ: Lawrence Erlbaum.

42.

Morgeson

F. P.

DeRue

D. S.

Karam

E. P.

(2010). Leadership in teams: A functional approach to understanding leadership structures and processes. Journal of Management, 36, 5-39.

43.

Muthén

L. K.

Muthén

B. O.

(1998-2012). Mplus user’s guide. Los Angeles, CA: Muthén & Muthén.

44.

Nicolaides

V. C.

LaPort

K. A.

Chen

T. R.

Tomassetti

A. J.

Weis

E. J.

Zaccaro

S. J.

Cortina

J. M.

(2014). The shared leadership of teams: A meta-analysis of proximal, distal, and moderating relationships. Leadership Quarterly, 25, 923-942.

45.

Nielsen

Daniels

(2016). The relationship between transformational leadership and follower sickness absence: The role of presenteeism. Work & Stress, 30, 193-208.

46.

Pearce

C. L.

(2004). The future of leadership: Combining vertical and shared leadership to transform knowledge work. Academy of Management Executive, 18(1), 47-57.

47.

Pearce

C. L.

Conger

J. A.

Locke

E. A.

(2008). Shared leadership theory. Leadership Quarterly, 19, 622-628.

48.

Pearce

C. L.

Ensley

M. D.

(2004). A reciprocal and longitudinal investigation of the innovation process: The central role of shared vision in product and process innovation teams (PPITs). Journal of Organizational Behavior, 25, 259-278.

49.

Perry

M. L.

Pearce

C. L.

Sims

H. P.

Jr. (1999). Empowered selling teams: How shared leadership can contribute to selling team outcomes. Journal of Personal Selling & Sales Management, 19(3), 35-51.

50.

Popper

Landau

Gluskinos

U. M.

(1992). The Israeli defense forces: An example of transformational leadership. Leadership & Organization Development Journal, 13, 3-8.

51.

Selig

J. P.

Preacher

K. J.

(2009). Mediation models for longitudinal data in developmental research. Research in Human Development, 6, 144-164.

52.

Shannon

H. S.

Robson

L. S.

Guastello

S. J.

(1999). Methodological criteria for evaluating occupational safety intervention research. Safety Science, 31, 161-179.

53.

Shikiar

Halpern

Rentz

Khan

(2004). Development of the Health and Work Questionnaire (HWQ): An instrument for assessing workplace productivity in relation to worker health. Work, 22, 219-229.

54.

Skinner

B. F.

(1963). Operant behavior. American Psychologist, 18(8), 503.

55.

Steenkamp

J. B. E.

Baumgartner

(1998). Assessing measurement invariance in cross-national consumer research. Journal of Consumer Research, 25(1), 78-90.

56.

Wang

Waldman

D. A.

Zhang

(2014). A meta-analysis of shared leadership and team effectiveness. Journal of Applied Psychology, 99, 181-198.

57.

Wang

I. S.

Courtright

S. H.

Colbert

A. E.

(2011). Transformational leadership and performance across criteria and levels: A meta-analytic review of 25 years of research. Group & Organization Management, 36, 223-270.

58.

Widaman

K. F.

Ferrer

Conger

R. D.

(2010). Factorial invariance within longitudinal structural equation models: Measuring the same construct across time. Child Development Perspectives, 4(1), 10-18.

59.

Selig

J. P.

Little

T. D.

(2013). Longitudinal data analysis. In Little

T. D.

(Ed.), The Oxford handbook of quantitative methods in psychology (Vol. 2., pp. 387-410). Oxford, England: University Press.