The Bright and Dark Sides of Distributed Leadership in Schools: A Joint Structural and Functional Perspective on Distributed Leadership, Work Performance and Job Satisfaction

Abstract

This study combines a structural and functional perspective on distributed leadership to disentangle its beneficial and detrimental effects on job satisfaction and work performance. Specifically, we explore the interaction between structural (SDL) and functional distributed leadership (FDL) on leadership support, organizational identification, and empowerment. This study also tests the mediating role of leadership support, organizational identification and empowerment as mechanisms that explain the association between distributed leadership and work-related outcomes in teachers. We used a multilevel mediation analysis to test the overall model in a sample of 2632 teachers embedded in 203 Romanian schools. The results replicate previous findings regarding the negative association between SDL and empowerment and identification and show that FDL has an overall positive association with leadership support, identification, and empowerment, as well as with job satisfaction and work performance reported by teachers. SDL had a negative indirect association with job satisfaction mediated by leadership support and with work performance mediated by organizational identification. The association between FDL and job satisfaction was significantly mediated by leadership support, identification, and empowerment within schools. Finally, the association between FDL and work performance was significantly mediated by organizational identification within as well as between schools.

1. Introduction

Distributed leadership is one of the most prolific research streams on academic leadership (Eryilmaz & Sandoval-Hernandez, 2023; Mifsud, 2023) and the vast majority of studies reveal positive implications for work-related and academic outcomes in schools (Liu & Werblow, 2019; Printy & Liu, 2021), while some studies show a negative association between distributed leadership and teachers’ job satisfaction and empowerment (Tucaliuc et al., 2023). In its early years, the concept of distributed leadership was marked by confusion (Harris, 2007) and, to date, the literature on distributed leadership still lacks conceptual clarity, as various studies focus on different conceptualizations and definitions (Mifsud, 2023). Moreover, a competing stream of research that focuses predominantly on charismatic leaders (Mifsud, 2023) points out the beneficial role of (single) principal’s leadership on teacher and school outcomes (Wu & Shen, 2022), leaving the question of how distributed leadership relates to work outcomes in schools, unresolved (Tucaliuc et al., 2023). Whether they focus on leadership enacted by a single individual or leadership practices distributed among school employees, previous studies argue that leadership is a resource pool that employees draw from (Tepper et al., 2018), a motivational force that empowers followers (Muijs & Harris, 2003), as well as a source for social identification (Hogg et al., 2012). To better understand the implications of distributed leadership in schools, it is crucial to jointly explore multiple mechanisms that drive teachers’ work-related outcomes (Tucaliuc et al., 2023).

The present study expands the literature on distributed leadership by simultaneously taking a structural and a functional approach. On the one hand, we look at distributed leadership in schools, through a structural lens and we evaluate it as the formal allocation of academic leadership roles to more teachers within the school. While in principle, distributed leadership could also involve support staff, not only teachers, given the leadership roles allocation in the Romanian educational system in which only high school teachers can in principle assume formal leading roles, our research focuses exclusively on teachers. On the other hand, we look at distributed leadership through a functional lens and we evaluate it as the perception of how various leadership functions (Lakomski, 2008) are jointly fulfilled by teachers. The structural perspective on distributed leadership (SDL) defines leadership roles as central positions in the hierarchical structure of schools, while the functional perspective on distributed leadership (FDL) looks at how various leadership functions are fulfilled by multiple individuals, with or without a formal acknowledgment of status differences across the ones that fulfil these leadership functions and the rest of the teachers. The vast majority of empirical studies so far take a functional perspective on distributed leadership (Liu & Werblow, 2019; Mifsud, 2023; Printy & Liu, 2021) and assess the extent to which teachers participate in functions such as decision making, career and professional development of teachers, and instructional management. Such studies view distributed leadership as an emergent property of complex systems (Lakomski, 2008) and focus on participation in decision making, work design, and organization, aspects of work agency that are typically associated with positive outcomes. The formal allocation of leadership roles to several teachers in the school creates various hierarchical structures and generates different work dynamics, as teachers have to report to different lines of authority and may experience a lack of identification and engagement (Tucaliuc et al., 2023).

In our study, we aim to explore the distinctive and joint influence of SDL and FDL on teachers’ job satisfaction and work performance. First, we distinguish between leadership as a form of structural versus functional differentiation, while combining the structural view (formal allocation of leadership roles among several teachers) and the functional view (teachers’ joint participation in fulfilment of various leadership functions) on distributed leadership. This integration allows us to explore such complex relations between SDL and FDL and teachers’ school outcomes. Second, our study simultaneously considers motivational (empowerment), resource-driven (leadership support), and identity-related mechanisms that can explain the association between distributed leadership and work-related outcomes in teachers. We attempt to replicate the results reported by Tucaliuc et al. (2023) on SDL and also use distrust cognitive schema as a predictor in our model, as the impact of SDL and FDL is likely to be contingent on the level of distrust reported by teachers. As such, our paper reports one of the first empirical attempts to disentangle the bright and dark sides of distributed leadership, hypothesizing that SDL has a negative, while FDL has a positive association with perceived supervisory support, empowerment, and organizational identification that in turn impacts on job satisfaction and work performance in teachers. We also explore the interplay between SDL and FDL by testing the moderating role of FDL in the relation between SDL and leadership support, school identification, and empowerment.

2. Theory and Hypotheses

The literature on distributed leadership in schools is marked by conceptual and methodological ambiguity (Tian et al., 2015) and in order to better emphasize the contributions of our paper, we start by defining educational leadership and distinguishing between the structural and functional perspectives on distributed leadership. Leadership in educational settings “involves the identification, acquisition, allocation, coordination, and use of social, material, and cultural resources necessary to establish the conditions for the possibility of teaching and learning” (Spillane et al., 2001, p. 24). Leadership practices can be further categorized as macro-level leadership practices (focused on strategic issues, planning, and educational vision) and micro-level practices, focused on the oversight of daily operations in schools (Spillane et al., 2001). The literature on educational leadership and management distinguishes between single-leader models typically focused on leadership traits and styles embodied by educational leaders (Daniëls et al., 2019) and distributed leadership focusing on leadership as shared, emergent property in social systems that helps them cope with environmental and inner complexity (Bolden, 2011; Lakomski, 2008). Both streams of research depict leadership as a key factor for school and educational success (Mifsud, 2023) by fostering teachers’ identification with the school, their engagement and empowerment, as well as by securing and managing key resources for educational processes (Daniëls et al., 2019; Lakomski, 2008; Spillane et al., 2001). In their seminal conceptual work on school leadership, Spillane et al. (2001) build on a distributed cognition framework to argue that leadership practices in schools are enacted by multiple individuals and therefore, are distributed (or stretched) across actors and artefacts. Such practices can be formally allocated to one individual within the school (typically the school principal) or to several teachers, leading to a hierarchical differentiation between teachers, formally designated with leadership roles and those without such responsibilities. Formal leadership roles are social artefacts and the structural perspective on distributed leadership concerns the formal allocation of leadership responsibilities to specific social roles (principal, divisional principals, curriculum directors, etc.), representing a structural perspective on distributed leadership. On the other side, a functional perspective builds on the assumption that leadership practices can be naturally enacted by different teachers in the school (Spillane et al., 2001) and this form of social differentiation (different teachers fulfilling different leadership functions) does not necessarily overlap with the formal hierarchical structure of the school.

To better understand the difference between SDL and FDL, we focus on some of the key mechanisms that translate leadership practices into organizational outcomes. We first build on the tenets of the Social Identity Theory of Leadership (Hogg et al., 2012) to argue that leaders are prototypical images of the organization and as such they serve as targets of organizational identification for teachers. In particular, teachers tend to identify stronger with schools that score high on prestige and have a high optimal distinctiveness, as recognized as distinct and desirable social settings by others. Moreover, we build on the Person-Environment Fit Theory of Leadership (Lambert et al., 2012; Tepper et al., 2018) to argue that leaders facilitate followers’ access to key resources, therefore teachers seek support from their leaders in gaining access to work-related (financial, educational and material support) as well as personal (feeling appreciated and values, emotional support) resources, that allows them to perform well and maintain a good quality of life at work. Finally, drawing upon the Self-Determination Theory (Ryan & Deci, 2000), we argue that leadership creates conditions at work that facilitate the fulfilment of basic psychological needs such as connectedness, autonomy, and competence. In other words, educational leaders empower teachers to engage with their work in a meaningful way that allows them to fulfil their basic psychological needs and support their personal quest for self-actualization (Eyal & Roth, 2011).

To summarize, building on these theoretical frameworks, we posit that educational leaders serve as social identification targets for teachers, are key sources of support, and empower them to achieve their work-related and personal goals. It is our contention that these mechanisms explain the association between FDL and SDL on the one hand and the work-related outcomes on the other hand. We will subsequently elaborate on the distinction between SDL and FDL to specify different ways in which these two forms of distributed leadership drive the identification, empowerment, and social support provided to the teachers.

Previous research has documented a negative association between SDL and organizational identification, showing that teachers working in organizational hierarchies with multiple leaders tend to identify less strongly with their schools compared with teachers working in schools with a clear hierarchical structure led by a single principal (Tucaliuc et al., 2023). A key argument surrounding the relationship between SDL and identification is that multipolar hierarchical structures in schools (hierarchical organization with two or more leaders) dilute organizational distinctiveness, thereby weakening the organizational identification of the teachers. To date, the literature shows that the number of interpersonal relationships from which people are expected to receive social support has a non-linear association with perceived social support and wellbeing (Muntean et al., 2024; Telecan et al., 2024). As such, when too many educational leaders are expected to provide social support for teachers, it is possible that social support is actually diluted as compared with situations when a single leader provides the necessary support. Moreover, bifurcated organizational structures also create ambiguity with respect to the lines of authority and support provision, therefore we expect that SDL will also lead to decreased social support perceived by teachers. Previous empirical evidence also shows that SDL breeds decreased empowerment in teachers (Tucaliuc et al., 2023), therefore we expect that SDL dilutes leadership supportive and empowering practices, and we hypothesize the following:

Hypothesis 1.

Structurally distributed leadership (SDL) has a negative association with leadership support, organizational identification, and empowerment.

The functional perspectives on distributed leadership focus on emerging leadership practices in schools (Lakomski, 2008), irrespective of whether the leading roles are assigned to a single principal or to multiple individuals in the school. When multiple teachers enact leadership practices in a spontaneous and natural way, they also provide a larger pool of relational resources that followers can draw from. Moreover, participation in decision making ultimately increases identification with the school and elevates the sense of empowerment in teachers, thereby fostering their work engagement (Bogler, 2005; Park et al., 2020). Such arguments are in line with the Social Interdependence Theory (Deutsch, 1949) stating that broad participation in key decision-making processes reflects the supportive role of leadership, and increases the likelihood that teachers experience positive interdependence, consequently leading to enhanced empowerment and support. When teachers are expected to fulfil various leadership functions, including planning and decision making, they are likely to perceive positive interdependence and elevate their perception of support and empowerment. Therefore, we hypothesize the following:

Hypothesis 2.

Functionally distributed leadership (FDL) has a positive association with perceived leadership support, organizational identification and empowerment.

We previously argued that FDL does not necessarily overlap with formal organizational hierarchies in schools (Tucaliuc et al., 2023), therefore it is likely that SDL and perceptions of FDL co-exist also in schools led by a single principal. Given the opposing effects expected from SDL and FDL on leadership support, identification, and empowerment, we expect that emergent FDL alleviates the negative association between SDL and these variables. Therefore, we hypothesize the following:

Hypothesis 3.

FDL moderates the association between SDL on the one hand and leadership support, identification, and empowerment on the other hand, such that the SDL is detrimental only when FDL is low rather than high.

Previous research showed that the association between SDL and work-related outcomes reported by teachers was mediated by organizational identification and empowerment (Tucaliuc et al., 2023). Consequently, we aim to replicate these findings and test the mediating role of leadership support, organizational identification, and empowerment in the relation between SDL and FDL on the one hand, and job satisfaction and work performance on the other hand.

Hypothesis 4.

Leadership support, organizational identification, and empowerment mediate the association between distributed leadership (SDL and FDL) on the one hand and work performance and job satisfaction on the other hand.

3. Methods

3.1. Sample and Procedure

Our study is based on a large-scale investigation of factors related to wellbeing and performance in schools, pooling data from four waves of data collection across more than 400 Romanian schools. The survey was distributed online to teachers via school management and the final sample included in our study consisted of 2632 teachers (209 male) with an average age of around 43 years old (SD = 9.8). The final sample included teachers embedded in 203 schools as we only included the respondents that provided full data on all variables included in the study and we received more than five valid responses per school. The study was approved by the Scientific Council of the Babeș-Bolyai University, Cluj-Napoca, Romania, participation was voluntary, and respondents received no compensation for their participation in the survey.

3.2. Scales

Formally distributed leadership roles (SDL) were evaluated by asking participants how many persons fulfilled leadership functions in their school and it was coded as a dummy variable (one leader versus multiple leaders).

Distributed leadership functions (FDL) and agency were evaluated with a scale adapted from Jønsson et al. (2016) by asking teachers to rate the extent to which they participate in (1) developing the vision and the strategy of the school, (2) teacher performance assessment, (3) decisions related to the organization of educational activities, (4) planning the educational activities, (5) decisions related to resource allocation, and (6) in planning of professional development activities for teachers. The answers were recorded on a five-point Likert scale (1 = not at all to 5 = very much) and Cronbach’s alpha for this scale was 0.95 showing an excellent reliability of the scale.

Leadership support was evaluated with three items adapted from Muntean et al. (2022) which asked participants to answer the following behaviorally anchored rating items related to instrumental support (“If I encounter tasks related problems at work I get support from my supervisor”), emotional support (“If I encounter emotional related problems at work I get support from my supervisor”), and relational support (“If I encounter conflict-related problems with my coworkers I get support from my supervisor”) received from their leaders. Cronbach’s alpha for this scale was 0.91 showing a good reliability of the scale.

Organizational identification was evaluated using a single graphical item of organizational identification (Shamir & Kark, 2004) and participants were asked to rate the extent to which they identified overlaps with the organization (school). The single-item measure was previously used to assess organizational identification in schools (Tucaliuc et al., 2023) and other organizational contexts (Cremers & Curşeu, 2024).

Empowerment was evaluated with a single behaviorally anchored item that asked teachers to “Reflect on how leadership functions are exercised in your school and select the value that best describes your situation: 1 = is restrictive and directive (we receive directions, clear instructions that limit our freedom to act) to 7 = is empowering (gives us the power and autonomy to decide how to do our work)”. This item was previously used to assess empowerment in schools (Tucaliuc et al., 2023) as well as other organizational settings (Cremers & Curşeu, 2023).

Work performance was rated using the scale developed by Rousseau and Aubé (2014) focusing on generic performance metric in organizations such as achievement of performance goals, overall quantity of work, overall quality of work, and the fulfilment of agreed commitments. Cronbach’s alpha for this scale was 0.82 showing a good reliability of the scale.

Job satisfaction was evaluated using a single item adapted from previous research (Nagy, 2002; Tucaliuc et al., 2023) (“How satisfied are you with your current job?”) and the answers were recorded on a 5-point Likert scale ranging from “1 = very unsatisfied” to “5 = very satisfied”.

Distrust dysfunctional cognitive schema was evaluated using five items selected from the Schema Questionnaire (Young & Brown, 1998) adapted for the Romanian population (Curseu et al., 2000) and previously used to assess dysfunctional cognition in teachers (Tucaliuc et al., 2023). For the distrust dysfunctional schema, Cronbach’s alpha was 0.70 showing a sufficient reliability of the scale. Distrust was used as a control variable, building on the previous results reported by Tucaliuc et al. (2023).

4. Results

Means, standard deviations, and correlations are presented in

Table 1. Means, standard deviations, and correlations.

	Mean	SD	1	2	3	4	5	6	7	8	9
1. Gender	0.08	0.270	1
2. Age	43.03	9.81	0.05 *	1
3. Distrust	2.3491	0.92894	0.018	0.057 **	1
4. SDL	0.5015	0.50009	0.043 *	0.164 ***	0.043 *	1
5. FDL	4.0909	0.82804	−0.005	−0.117 ***	−0.199 **	−0.126 **	1
6. Organizational identification	5.95	1.025	0.001	0.017	−0.193 **	−0.068 **	0.377 **	1
7. Leadership support	5.9600	1.40038	0.004	−0.119 **	−0.212 **	−0.148 **	0.555 **	0.384 **	1
8. Empowerment	5.11	1.289	−0.005	−0.032	−0.165 **	−0.059 **	0.382 **	0.300 **	0.481 **	1
9. Job satisfaction	4.26	0.997	0.017	−0.090 **	−0.117 **	−0.061 **	0.275 **	0.301 **	0.288 **	0.237 **	1
10. Work performance	4.3951	0.50263	−0.071 **	0.018	−0.149 **	−0.060 **	0.293 **	0.381 **	0.217 **	0.162 **	0.232 **

Note: gender was dummy coded Female = 0, Male = 1; SDL = structural distributed leadership was dummy coded 0 = one leader, 1 = more leaders; FDL = functional distributed leadership. *** p < 0.001 ** p < 0.01, and * p < 0.05.

.

Because our data were collected from the same source, we performed a Confirmatory Factor Analysis using JASP version 0.19.3.0, in order to check the validity of our measurement model. The model with five correlated factors (we included the single-item measures in the same factor) had a good fit with the data χ² = 1704.02, df = 220, p < 0.001, CFI = 0.96, TLI = 0.95, NFI = 0.95, RMSEA = 0.05, AIC = 139,639.91, and BIC = 139,968.94 while the single-factor solution did not fit the data well (χ² = 23,376.26, df = 209, p < 0.001, CFI = 0.62, TLI = 0.58, NFI = 0.62, RMSEA = 0.15), AIC = 255,944.50, and BIC = 256,228.44), as the difference between models was significant: ∆χ² = 21,672.24, ∆df = 9, p < 0.00001. We can therefore conclude that the measurement model was supported and the results are not likely to be overestimated due to the common method bias. Our data were collected from teachers nested in schools, therefore we analysed our data using multilevel analyses and used a macro developed for SPSS 29 by Rockwood and Hayes (2017). The multilevel analyses using the MLMed macro distinguished within from between school effects and allow for a parsimonious test of our hypotheses. In order to test the hypothesized main effects and the moderation hypotheses, we ran separate models for leadership support, organizational identification, and empowerment starting with a simple model that included gender, age, distrust, and SDL as predictors (Model 1 in

Table 2. Results of the multilevel analyses.

Variable	Leadership Support				Organizational Identification				EMPOWERMENT
Model	Model 1		Model 2		Model 1		Model 2		Model 1		Model 2
Level	Within Schools	Between Schools	Within Schools	Between Schools	Within Schools	Between Schools	Within Schools	Between Schools	Within Schools	Between Schools	Within Schools	Between Schools
Constant	6.10 *** (0.06)		6.01 *** (0.05)		6.01 *** (0.04)		5.97 *** (0.03)		5.17 *** (0.06)		5.11 *** (0.05)
Gender	0.09 (0.10)	0.29 (0.43)	0.04 (0.09)	0.48 (0.32)	0.04 (0.08)	−0.08 (0.26)	0.02 (0.07)	0.02 (0.24)	−0.01 (0.09)	0.06 (0.40)	−0.04 (0.09)	0.20 (0.36)
Age	−0.01 ***	−0.01	−0.007 **	−0.003	0.005	0.001	0.008 ***	0.005	−0.003	0.002	0.001	0.008
	(0.003)	(0.01)	(0.003)	(0.007)	(0.002)	(0.006)	(0.002)	(0.005)	(0.003)	(0.01)	(0.003)	(0.008)
Distrust	−0.27 ***	−0.52 ***	−0.14 ***	−0.25 **	−0.20 ***	−0.29 ***	−0.13 ***	−0.18 **	−0.21 ***	−0.29 **	−0.120 *** (0.03)	−0.13
	(0.03)	(0.12)	(0.03)	(0.09)	(0.02)	(0.07)	(0.02)	(0.07)	(0.03)	(0.11)	(0.03)	(0.10)
SDL		−0.36 ***		−0.16 **		−0.11 *		−0.04		−0.15 *		−0.03
		(0.08)		(0.06)		(0.05)		(0.05)		(0.08)		(0.07)
FDL			0.77 ***	1.09 ***			0.48 ***	0.49 ***			0.52 ***	0.67 ***
			(0.04)	(0.10)			(0.03)	(0.07)			(0.04)	(0.11)
SDLXFDL				0.12 * (0.06)				−0.07 † (0.05)				0.04 (0.06)

Note: unstandardized regression coefficients are shown with robust standard errors between parentheses; gender was dummy coded Female = 0, Male = 1; SDL = structural distributed leadership was dummy coded 0 = one leader, 1 = more leaders; FDL = functional distributed leadership. *** p < 0.001 ** p < 0.01, and * p < 0.05. ^† p < 0.10.

) and then added the FDL and the interaction effect between SDL and FDL in Model 2 (see Table 2). The results of the multilevel analyses are presented in Table 2.

As illustrated in Model 1, SDL has a negative and significant between schools association with leadership support (B = −0.36, SE = 0.08), with organizational identification (B = −0.11, SE = 0.05), and with empowerment (B = −0.15, SE = 0.08). This pattern of results replicates the negative effects of SDL reported in Tucaliuc et al. (2023) and fully supports Hypothesis 1. In Model 2, after adding FDL and the interaction term between SDL and FDL, the main effects of SDL become not significant, with the exception of the main effect of SDL on leadership support that remained negative and significant (B = −0.16, SE = 0.06, p = 0.01). Such a pattern of results shows that FDL and the interaction between FDL and SDL qualify the main effects of SDL.

As indicated in Model 2, FDL has a positive and significant association with leadership support within (B = 0.77, SE = 0.04) as well as between schools (B = 1.09, SE = 0.10). FDL has a positive and significant association with organizational identification within (B = 0.48, SE = 0.03) as well as between schools (B = 0.49, SE = 0.07) as well as a positive and significant association with empowerment within (B = 0.52, SE = 0.04) as well as between schools (B = 0.67, SE = 0.11). These results fully support Hypothesis 2 showing that contrary to SDL, FDL has a positive association with leadership support, organizational identification, and empowerment. Moreover, the interaction effect between SDL and FDL was only significant for leadership support (B = 0.12, SE = 0.06) and marginally significant for organizational identification (B = −0.07, SE = 0.05, p = 0.09) providing only partial support for Hypothesis 3. The significant interaction effect is presented in Figure 1, showing that the perceived leadership support in schools led by a single leader is significantly higher than in schools led by more leaders only when FDL is low, rather than high. Despite this moderation effect being consistent with Hypothesis 3, the hypothesis received only partial support.

The indirect effects were estimated between schools for SDL and within and between schools for FDL. The effect of SDL on job satisfaction was significantly mediated by leadership support (indirect effect −0.05, SE = 0.02, p = 0.01, 95% CI [−0.10; −0.02]), while the indirect effect of SDL on job satisfaction via identification (indirect effect −0.02, SE = 0.01, p = 0.15, 95% CI [−0.05; −0.01]) and empowerment (indirect effect −0.002, SE = 0.009, p = 0.88, 95% CI [−0.02; 0.02]) were not significant as the 95% CI included zero. The effect of SDL on work performance was significantly mediated by organizational identification (indirect effect −0.02, SE = 0.01, p = 0.06, 95% CI [−0.04; −0.002]) as the 95% CI did not include zero. The indirect effects of SDL on work performance via empowerment and leadership support were not significant. Given these results, Hypothesis 4 received only marginal support for SDL.

The association between FDL and work performance was significantly mediated by organizational identification (within schools indirect effect 0.07, SE = 0.006, p < 0.001, 95% CI [0.06, 0.08] and between schools indirect effect 0.07, SE = 0.02, p = 0.001, 95% CI [0.03; 0.12]). Finally, the association between FDL and job satisfaction was significantly mediated within schools by all mediators (indirect effect via organizational identification 0.09, SE = 0.01, p < 0.001, 95% CI [0.07; 0.11], via leadership support 0.07, SE = 0.02, p < 0.001, 95% CI [0.04; 0.09] and via empowerment 0.04, SE = 0.01, p < 0.001, 95% CI [0.02; 0.06]). None of the indirect paths between FDL and job satisfaction were significant when estimated between schools. We can therefore conclude that Hypothesis 4 was only supported for the within schools indirect effects of FDL.

In addition to the hypothesized relations, our study also replicated the negative association between distrust cognitive schema and organizational identification and empowerment reported in previous research (Tucaliuc et al., 2023). Distrust also has a negative association with leadership support in all models reported in

Table 3. Results of the multilevel mediation analyses.

Model	Job Satisfaction		Work Performance
	Within Schools	Between Schools	Within Schools	Between Schools
Constant	4.25 *** (0.03)		4.43 *** (0.02)
Gender	0.08 (0.07)	0.06 (0.23)	−0.11 ** (0.03)	−0.32 * (0.13)
Age	−0.006 ** (0.002)	−0.008 (0.005)	0.002 (0.001)	0.004 (0.003)
Distrust	−0.03 (0.02)	0.06 (0.07)	−0.03 ** (0.01)	−0.03 (0.04)
SDL		−0.004 (0.05)		−0.02 (0.03)
FDL	0.12 ** (0.04)	0.340 *** (0.09)	0.11 *** (0.02)	0.20 *** (0.05)
Leadership support	0.08 *** (0.02)	0.04 (0.06)	0.007 (0.009)	−0.06 † (0.03)
Organizational identification	0.20 *** (0.02)	0.14 † (0.08)	0.15 *** (0.01)	0.16 *** (0.04)
Empowerment	0.08 *** (0.02)	−0.009 (0.05)	−0.003 (0.008)	0.004 (0.03)

Note: unstandardized coefficients are shown with standard errors between parentheses; gender was dummy coded Female = 0, Male = 1; SDL = structural distributed leadership was dummy coded 0 = one leader, 1 = more leaders; FDL = functional distributed leadership; *** p < 0.001 ** p < 0.01, and * p < 0.05. ^† p < 0.10.

. Such results reflect the detrimental effect of distrust cognitive schema in work contexts. Moreover, age has a negative and significant association with leadership support, organizational identification, and job satisfaction, revealing that older teachers tend to report less leadership support and weaker identification with the school and are ultimately less satisfied with their jobs than younger employees. Finally, gender has a negative association with work performance, such that male teachers report lower levels of performance as compared to female teachers. This result is likely an artefact of the unbalanced gender composition of our sample, a situation that reflects the overall gender composition of the work force in the Romanian educational system (Muntean et al., 2022; Tucaliuc et al., 2023).

5. Discussion

Our study aimed to disentangle the detrimental and beneficial effects of distributed leadership in schools. By simultaneously using structural and functional perspectives on distributed leadership, we showed that when distributed leadership is conceptualized as a feature of school hierarchical organizational structures, its effects are predominantly negative. Specifically, leadership support, organizational identification, and empowerment were lower in schools led by multiple formal leaders compared to those led by a single principal. These results replicate previous findings on SDL (Tucaliuc et al., 2023) showing that when distributed leadership reflects a split in the hierarchical differentiation in schools (two or more power poles co-exist) the motivational, identity, and support functions of leadership are diluted. When taking a functional perspective on distributed leadership and treating it as an emergent, potentially self-organizing feature of a complex social system, its effects on identification, empowerment, and leadership support are predominantly positive. These opposing results, regarding the role of distributed leadership, raise awareness for the need to use an accurate conceptualization of this particular leadership form in educational settings.

A key contribution of our paper is the examination of the interplay between SDL and FDL on empowerment, leadership support, and organizational identification. Only the interaction effect on leadership support was significant, showing that perceived leadership support covaries negatively with SDL only when FDL is low. The interaction effect between SDL and FDL on organizational identification was marginally significant (as such we do not interpret this as a significant moderation), while the effect on empowerment was not significant. These results did not offer substantial support for our expected moderation. A plausible explanation is the negative and significant association between SDL and FDL showing that the number of formal leaders in the school triggers lower levels of perceived FDL. In other words, schools in which two or more teachers have formal leading roles, tend to score lower on FDL. This negative association may have affected the interaction effects hypothesized in our study and is somewhat counterintuitive as it would be expected that multiple leadership roles would facilitate the distribution of leadership functions among teachers in the schools.

The FDL assessment used in our study reflects an agency view on distributed leadership, meaning that the items of the scale truly reflect the extent to which teachers in our sample perceive they are involved in the decision-making process at different levels of the educational process (from curricula development to planning and monitoring educational processes). Such a leadership agency captures distributed leadership as an emergent property (potentially disconnected from the formal school hierarchy) of schools and embodies the nature of participation for work-related attitudes and work performance. Teachers in schools led by a single principal tend to report higher levels of FDL, meaning that when formal leading roles are allocated to multiple teachers, they may lack the knowledge and expertise to really empower the other teacher to engage in leadership practices, or there is a saturation concerning the leadership functions left to be fulfilled emergently by other school employees. This result is paramount, showing that a simple allocation of a formal leading role to different teachers within the schools is not, in and of itself, beneficial for the support these leaders provide to the teachers. More than this, formally distributing leadership functions to (unprepared) teachers may deprive others of the opportunities of experiencing leadership, personal development, or even fulfilling their leadership potential. We join the voices calling for formal leadership training offered to educational managers in order to support them in dealing with the complexity of their leading roles (Tucaliuc et al., 2023, 2024).

Our study tackles yet another tension point between distributed leadership and a more traditional leadership perspective on academic management (focused on a single leader), namely the mechanisms that explain the association between school leadership and work outcomes. To date, the literature agrees that leaders as prototypical images of the organization, provide resources, foster social identification, and are sources of empowerment and work engagement for teachers (Tucaliuc et al., 2023). These are all mechanisms through which hierarchical as well as distributed leadership in schools affect work-related outcomes. Although our results show that single prototypical leaders trigger stronger identification tendencies, empower, and support, when the agentic view on distributed leadership is accounted for, this superiority effect of a single prototypical leadership is overshadowed. Therefore, the extent to which teachers perceive they have agency in performing various leadership tasks ultimately drives their identification tendencies, perceptions of empowerment, and support. These results show that it is paramount to train educational leaders to become more instrumental in effectively delegating leadership practices within their schools and supporting their staff to engage more effectively in the decision-making processes.

6. Limitations and Avenues for Further Research

Next, we report several limitations of our study. First, our design was cross-sectional and although we used multilevel analyses, and the CFA reduces concerns related to the common method bias, we cannot derive clear causal claims based on our results. Moreover, as we tested moderation effects, in line with previous simulation results (Siemsen et al., 2010) we could expect that the interaction effects were underestimated, leading to marginally significant effects for organizational identification. Future research could further explore this interaction effect in other samples. In addition, it is unlikely that the significant interaction effect between SDL and FDL on leadership support was overestimated due to common method bias (Siemsen et al., 2010). Second, we used single-item measures for identification, empowerment, and job satisfaction and although we rely on existing single-item measures that comply with the requirements stated in the literature for such measures (Allen et al., 2022), we have to acknowledge the limitations related to the validity of our single-item measures. Third, it is important to acknowledge the use of an agency-based measure of distributed leadership in schools and to note that alternative measures of distributed leadership may lead to different effects. Nevertheless, we call for more integration in the conceptual refinement of distributed leadership, in order to clarify the methodological constraints that would advance the state of the art in distributed leadership research. Finally, our study was cross-sectional, therefore no causal claims can be made based on the results reported here. It is difficult to really manipulate FDL, yet future studies could explore in cvasi-experimental designs the way in which (manipulated) SDL and emergent forms of FLD impact on various outcomes in small groups. Also, future research using longitudinal designs could more clearly disentangle the role of supervisory support, organizational identification, and empowerment as mediators.

7. Practical and Policy Implications

Distributed leadership dominates the literature on educational management, leadership, and administration becoming almost normative for school administrators (Mifsud, 2023). Such normativity raises important implementation challenges, as educational administrators tend to use distributed leadership loosely and indiscriminately. School administrators should carefully design organizational structures in order to avoid the fragmentation of hierarchical structures that would eventually weaken leadership support, organizational identification, and reduce teacher empowerment. As our results show, simply assigning multiple leading roles in schools does not guarantee improved leadership outcomes, therefore we join voices (Eryilmaz & Sandoval-Hernandez, 2023; Tucaliuc et al., 2023, 2024) calling for leadership training for teachers as well as for educational managers in order to facilitate the effective implementation of distributed leadership. We are aware that, especially in large schools, it is paramount to allocate formal leading roles to more teachers yet based on our results we advise against leadership dilution and for a clear definition and allocation of leadership responsibilities in order to support teacher empowerment. A second set of practical implications refers to increasing teacher agency and emphasizing a functional rather than structural approach to distributed leadership. Educational administrators should create a safe space for participation in which emergent self-organized leadership practices naturally occur and evolve, ultimately improving school outcomes.

In terms of policy implications, our results call for mandating professional development programmes for educational leaders, in order to equip them with the skills to delegate leadership responsibilities effectively and to implement participatory practices that enhance teacher agency in school decisions. A particular policy implication refers to incorporating school assessment systems that take into account teacher agency, making sure that educational leadership genuinely facilitates participation in decision making. Such policies imply that single-leader educational approaches could be supplemented with participatory work design features, in such a way that even in cases in which schools are led by a single principal, the teachers can still engage in leadership practices, supporting overall educational processes.

8. Conclusions

In conclusion, the results highlight the importance of considering a more nuanced understanding of distributed leadership in schools, given the differential impact of SDL and FDL on leadership support, organizational identification and empowerment, and ultimately job performance and satisfaction. These findings offer valuable contributions to both theoretical approaches and practical applications in (distributed) educational leadership. Moreover, they pave the way for further investigation into the types, forms or models of school leadership that best account for educational outcomes.