Promoting Inclusive Education Through School Leadership: The Mediating Role of Principal Self-Efficacy

Abstract

Inclusive education depends not only on policy frameworks but also on school leaders’ capacity to translate inclusive principles into sustained organisational practice. This study examined the relationships among principals’ transformational and transactional leadership, self-efficacy, school culture, and the implementation of inclusive policies and practices. The study employed a cross-sectional quantitative design involving 452 principals of primary and lower-secondary schools in Greece. Data were collected using the Leadership for Inclusion Questionnaire (LEI-Q), the School Culture Survey (SCS), the Principals’ Self-Efficacy for Inclusive Leadership Scale, and the Transformational and Transactional Leadership Scale (TATL). Both leadership styles were positively associated with school culture, although transformational leadership showed stronger and more consistent relationships. Hierarchical regression explained 50% of the variance in inclusive leadership practices. Principals’ self-efficacy accounted for 45% of the explained variance, while transformational leadership contributed an additional 4%. Mediation analyses revealed significant indirect effects of both leadership styles through self-efficacy across all four dimensions of inclusive policy and practice implementation. The findings identify principals’ self-efficacy as the principal mechanism through which leadership orientations are translated into inclusive action, highlighting the importance of strengthening leaders’ confidence and capacity alongside organisational conditions that support sustainable inclusive school development.

1. Introduction

Inclusive education has become a central priority of contemporary educational policy, reflecting international commitments to equity, participation, and the right of every learner to quality education (Yang et al., 2025). Contemporary frameworks conceptualize inclusion not simply as the placement of students with disabilities in mainstream settings but as a systemic process of educational transformation that seeks to remove barriers to learning and participation for all learners, regardless of disability, ethnicity, language, gender, or socioeconomic background (Graham, 2020; Graham et al., 2023; UNESCO, 2020; Zoniou-Sideri, 2018). This understanding is consistent with the United Nations Committee on the Rights of Persons with Disabilities (General Comment No. 4) (UN, 2016), which defines inclusive education as a systemic process of educational transformation involving modifications in curricula, teaching methods, organizational structures, and educational policies to eliminate barriers to learning and participation (Graham, 2020). Consequently, inclusive education extends beyond technical or administrative adjustments and requires schools to reconsider institutional structures, pedagogical practices, and deeply embedded assumptions about normality, achievement, ability, and belonging (Slee, 2011, 2019; Vlachou, 2023).

Despite this international consensus, translating inclusive policy into everyday educational practice remains a persistent global challenge. Educational systems continue to operate within broader social, political, and economic structures that often reproduce rather than reduce inequalities (Bakogiannis & Papavasiliou, 2026). Poverty, geographical disparities, migration, cultural diversity, unequal access to specialist support services, and limited educational resources substantially influence schools’ capacity to implement inclusive policies effectively (Graham, 2020; Tsirantonaki & Vlachou, 2024b; Vlachou, 2023). Moreover, exclusion is not solely produced through material inequalities. It is also embedded within institutional cultures, professional beliefs, accountability systems, and long-established assumptions about academic performance and student ability (Graham, 2020; Slee, 2014; Spandagou, 2021). As Slee (2018) argues, educational exclusion is not an accidental by-product of schooling but is historically embedded within the very structures through which schooling has traditionally been organized.

These complexities suggest that inclusive education cannot be understood simply as the implementation of policy directives (Hardy & Woodcock, 2024). Rather, inclusion is negotiated within schools themselves, where educational policies are interpreted, adapted, and enacted through everyday interactions among school leaders, teachers, students, and families (Ainscow, 2020; DeMatthews et al., 2020; Oskarsdottir et al., 2020). Consequently, the school represents the critical micro-level at which the success or failure of inclusive reform is ultimately determined (Rapp & Corral-Granados, 2024).

Within this micro-context, school leadership has emerged as one of the strongest organizational determinants of successful inclusion. Contemporary research consistently demonstrates that principals influence not only administrative processes but also the professional norms, collaborative practices, and organizational cultures that shape teachers’ work and students’ educational experiences (Berkovich & Eyal, 2017, 2018; Pazey & Combes, 2020; Tsirantonaki & Vlachou, 2024a). Inclusive leadership therefore extends beyond managerial efficiency; it involves establishing a shared vision grounded in equity, fostering collaborative professional learning, promoting distributed responsibility, and creating organizational conditions that support participation for all students (Carrington, 2022; Crisol-Moya et al., 2020).

Equally important is principals’ perceived self-efficacy. Drawing on Bandura’s (1997) social cognitive theory, leadership self-efficacy refers to leaders’ beliefs in their capability to organize and execute the actions necessary to achieve desired organizational outcomes. Within educational settings, principal self-efficacy has been associated with greater persistence in addressing complex problems, stronger commitment to school improvement, higher levels of innovation, and greater willingness to implement demanding educational reforms (Schmitz & Tulowitzki, 2025; Röhl et al., 2024; Yilmaz & Tore, 2025). In inclusive education specifically, leaders with stronger efficacy beliefs appear more willing to introduce differentiated instructional approaches, support collaborative teaching, redesign school policies, and promote equitable participation across diverse student populations (Tsiolpidou & Koutouzis, 2024; T. Wang et al., 2022).

Although the educational leadership literature has consistently acknowledged the importance of school principals in advancing inclusive education, the empirical evidence remains conceptually fragmented. Previous research has predominantly examined inclusive leadership, leadership styles, principal self-efficacy, and school culture as separate or only pairwise constructs, demonstrating positive associations among these variables but offering limited insight into the mechanisms through which they collectively shape inclusive school development (Berkovich & Eyal, 2017, 2018; Tschannen-Moran & Gareis, 2004). As a result, current knowledge explains whether leadership matters considerably better than it explains how leadership translates into the implementation of inclusive policies and practices within schools.

In particular, relatively little attention has been devoted to examining the simultaneous predictive contribution of principals’ self-efficacy, leadership styles, and school culture within a single explanatory framework. While transformational leadership has repeatedly been associated with positive organizational outcomes and inclusive school improvement (Crisol-Moya et al., 2020; DeMatthews et al., 2020; Khaleel et al., 2021), considerably less is known about its relative contribution compared with principals’ beliefs in their own leadership capabilities. Likewise, transactional leadership is frequently portrayed as less compatible with inclusive reform, yet empirical evidence regarding its role in shaping inclusive school practices remains inconclusive (Hieng et al., 2024; Maurice et al., 2024). More importantly, previous studies have rarely investigated whether leadership influences inclusive school development directly or indirectly through principals’ self-efficacy (Lambrecht et al., 2020). Consequently, the processes linking leadership behaviours, leaders’ efficacy beliefs, and the implementation of inclusive policies remain insufficiently understood.

Addressing these limitations requires explanatory models capable of identifying both the relative contribution of leadership variables and the mechanisms through which leadership is translated into inclusive practice. Understanding these pathways is essential because inclusion is enacted through everyday school leadership as principals navigate the often contradictory and complex policy environments within which inclusive practice is developed (Koullapi & Lyra, 2026).

Against this background, the purpose of this study is to examine how principals’ leadership styles and perceived self-efficacy contribute to the implementation of inclusive school policies and practices. Specifically, this study examines the relationships among leadership styles, school culture, principal self-efficacy, and inclusive leadership practices; evaluates their relative predictive contribution; and tests whether principal self-efficacy mediates the relationship between leadership styles and four dimensions of inclusive policy and practice implementation: Inclusive Leadership and School Culture, which reflects leadership practices aimed at developing shared inclusive values and organisational conditions; School–Community Inclusive Policies, concerning the development of inclusive policies and collaboration between the school and its wider community; Inclusive Teaching and Assessment Practices, referring to organisational support for equitable and responsive approaches to teaching, learning, and assessment; and Student and Family Empowerment, which captures practices that promote the participation and involvement of students and families in school life.

The present study makes several contributions to the literature. First, it develops and empirically tests an integrated explanatory framework that simultaneously examines principals’ self-efficacy, transformational and transactional leadership, school culture, and inclusive leadership practices, rather than considering these constructs in isolation. By doing so, the study responds to growing calls for more comprehensive models capable of explaining the organizational processes underlying inclusive school development (DeMatthews et al., 2020; Graham, 2020; Mavrou et al., 2025).

Second, the study moves beyond traditional correlation analyses by investigating both prediction and mechanism. Using hierarchical multiple regression, it examines the relative contribution of principals’ self-efficacy, leadership styles, and school culture in predicting the implementation of inclusive policies and practices. Furthermore, by employing mediation analysis using Hayes’ PROCESS procedure for SPSS, the study examines whether principals’ self-efficacy functions as the psychological mechanism through which transformational and transactional leadership influence the implementation of inclusive practices across multiple dimensions of inclusive school development. This analytical approach provides a substantially more nuanced understanding of leadership processes than direct-effect models alone (Hayes & Rockwood, 2017).

2. Theoretical Background

2.1. Inclusive School Culture

Inclusive school culture shifts attention from the placement of students within existing structures to the transformation of schools as organisations (Hansen et al., 2020). Creating an inclusive school culture is a complex and ongoing process that extends far beyond the implementation of inclusive policies or the provision of additional support services. It requires a fundamental shift in the values, beliefs, and everyday practices that shape teaching, learning, and interpersonal relationships within the school community (Ainscow, 2020; Slee, 2018). Inclusive cultures are developed through continuous negotiation among multiple stakeholders—including school leaders, teachers, students, families, and external agencies—whose perspectives, expectations, and capacities may not always align. This complexity is further influenced by contextual factors such as resource availability, accountability demands, professional learning opportunities, and broader sociocultural attitudes toward diversity and disability (Graham, 2020; Vlachou, 2023). Consequently, fostering inclusion involves addressing organisational structures alongside the underlying norms and assumptions that determine who belongs, whose voices are heard, and how diversity is understood and valued (Spandagou, 2021; Tsirantonaki & Vlachou, 2024b). Rather than representing a fixed organisational state, an inclusive school culture is a dynamic and evolving process that depends on sustained collaboration, reflective practice, and leadership capable of balancing competing priorities while maintaining a clear commitment to equity and participation for all learners (Carrington, 2022; DeMatthews et al., 2020).

Such culture cannot be separated from the political and ideological conditions within which schooling operates. Slee (2018) positions inclusion as a political project that unsettles dominant assumptions about normality, ability, and achievement, while Graham (2020) cautions against reducing it to procedural compliance. Formal policies may proclaim inclusion while everyday routines continue to legitimise exclusion. What matters, therefore, is whether equity-oriented values become embedded in the collective assumptions, institutional practices, and pedagogical decisions of the school community.

School culture generally refers to the shared beliefs, values, norms, and assumptions that shape organisational behaviour and relationships (Cucchiara et al., 2025; Gruenert & Whitaker, 2021). Inclusive school culture, however, demands more than collaboration, trust, or organisational effectiveness. It requires diversity to be recognised as an educational resource and participation to be treated as a matter of justice rather than administrative accommodation (Carrington, 2022). Schools characterised by shared responsibility, high expectations for all learners, and collaborative professional relationships are more likely to sustain inclusive practices (Ainscow, 2020; Vlachou, 2023). Conversely, cultures structured by deficit thinking, rigid accountability, and assumptions of homogeneity may reproduce exclusion even while claiming formal commitment to inclusion (Slee, 2018).

Leadership plays a central role in shaping school culture, as principals may influence not only organisational arrangements but also what is valued, normalised, and ultimately enacted in everyday school practice (DeMatthews & Mawhinney, 2014; Gu, 2023). Fostering inclusion, however, is one of the most demanding responsibilities of school leadership because it requires principals to lead change across multiple and interconnected dimensions of school life. Beyond supporting the implementation of inclusive education policies, principals are expected to cultivate a shared vision, build staff commitment, address resistance to change, and create organisational conditions that support equitable teaching and learning (Carrington et al., 2024; DeMatthews et al., 2021). This work is often complicated by competing accountability pressures, limited resources, diverse student needs, and variations in teachers’ beliefs, confidence, and preparedness to implement inclusive practices (Graham, 2020; Hardy & Woodcock, 2024; Vlachou, 2023). At the same time, principals must balance administrative responsibilities with instructional leadership, coordinate collaboration among teachers and specialist professionals, engage families and community partners, and advocate for additional support where necessary (DeMatthews et al., 2020; Valdés et al., 2026). Because school cultures evolve gradually, fostering inclusion requires sustained leadership that promotes trust, collective responsibility, and continuous professional learning rather than isolated initiatives (Carrington, 2022; Gu, 2023). Consequently, successful inclusive leadership depends not only on principals’ commitment and capabilities but also on the organisational and systemic conditions that enable meaningful and lasting change (Graham, 2020; Slee, 2018; Tsirantonaki & Vlachou, 2024b).

Despite the recognised importance of school leadership, considerably less attention has been paid to the psychological processes through which leadership is translated into sustained inclusive practice. In particular, principal self-efficacy has received limited attention as a potential mechanism linking leadership orientations with the implementation of inclusive policies and practices across complex organisational settings (Liu, 2024; T. Wang et al., 2022). The following section therefore examines principal self-efficacy and its theoretical relevance to inclusive educational leadership.

2.2. Principal Self-Efficacy

Principal self-efficacy, grounded in Bandura’s Social Cognitive Theory, refers to school leaders’ beliefs in their capability to organise and execute the actions necessary to achieve desired educational outcomes (Bandura, 1997, 2012). Unlike general self-confidence, self-efficacy is context-specific and influences how individuals interpret challenges, regulate effort, persist in the face of obstacles, and respond to adversity. Within educational leadership, it reflects principals’ beliefs that they can successfully mobilise people, resources, and organisational processes to support school improvement (Federici & Skaalvik, 2011; Tschannen-Moran & Gareis, 2004).

Research consistently associates stronger efficacy beliefs with greater persistence, professional engagement, innovation, and capacity for organisational change (Skaalvik, 2020; Federici & Skaalvik, 2011). In inclusive education, these beliefs appear particularly important because principals are required to challenge established assumptions about ability, coordinate collaboration among multiple stakeholders, and sustain reform under conditions of uncertainty and resistance. Inclusive leadership self-efficacy has therefore been conceptualised as principals’ confidence in their ability to create equitable learning environments, promote collaboration, and support the participation and success of all learners (Rivana et al., 2024; Tsiolpidou & Koutouzis, 2024; T. Wang et al., 2022).

At the same time, self-efficacy should not be interpreted as an individualistic explanation for successful leadership. As Slee (2011) argues, educational change is shaped not only by leaders’ beliefs but also by institutional conditions, accountability systems, available resources, and broader policy environments. High self-efficacy alone cannot overcome structural barriers to inclusion. Rather, efficacy beliefs should be understood as relationally produced and enacted within specific organisational and political contexts (Hamann et al., 2024; Mustafa et al., 2019).

Empirical evidence nevertheless suggests that self-efficacy plays an important role in inclusive school development. Principals with stronger efficacy beliefs report greater willingness to implement inclusive practices, strengthen collaboration with teachers and families, and promote school cultures characterised by participation and equity (Tsiolpidou & Koutouzis, 2024; Van Staden-Payne & Nel, 2023). Self-efficacy has also been linked to transformational leadership, organisational learning, collective teacher efficacy, and innovation (Gümüş et al., 2026; Zhang et al., 2026). Consequently, principal self-efficacy may represent a key mechanism through which leadership orientations are translated into inclusive school practices (T. Wang et al., 2022), functioning both as a predictor of inclusive action and as a potential mediator linking leadership styles and the promotion more inclusive learning environments.

Importantly, the translation of leadership self-efficacy into inclusive action is conditioned by the institutional environments in which principals work. Accountability and performativity pressures may constrain professional discretion by prioritising measurable attainment and procedural compliance, while unequal access to staffing, specialist expertise, time, and material resources may limit what even highly efficacious leaders are able to enact. Self-efficacy should therefore be understood not as an autonomous individual resource, but as a form of agency whose practical expression is enabled, constrained, and continually reshaped by organisational and policy conditions (Graham, 2020; Hardy & Woodcock, 2024; Slee, 2011; Vlachou, 2023).

2.3. Transformational and Transactional Leadership: Between Transformation and Regulation in Inclusive Education

Among the many leadership approaches discussed in educational research, transformational and transactional leadership remain particularly relevant to inclusive education because they represent two distinct ways of responding to organisational change, diversity, and school improvement (Leithwood & Jantzi, 2006; Li & Liu, 2020). Rather than treating leadership as a neutral managerial tool, a critical perspective views leadership practices as embedded within broader questions of power, participation, and educational justice (Slee, 2018, 2019).

Transformational leadership emphasises the development of a shared vision, intellectual stimulation, individualised support, and collective commitment (Bass & Avolio, 1997; Erden & Yaşlioğlu, 2020; Leithwood & Jantzi, 2006). These dimensions appear particularly relevant to inclusion because they encourage schools to reconsider what is valued, whose voices are recognised, and how participation is organised (Carrington et al., 2024). Empirical evidence consistently links transformational leadership to positive school cultures, teacher commitment, professional collaboration, and inclusive practices (Z. Wang et al., 2026; Wilson Heenan et al., 2023). However, critical scholars caution against presenting transformational leadership as a universal solution (Huang et al., 2026). Vision and commitment alone do not remove structural inequalities, and leadership discourse can become detached from the institutional conditions that continue to produce exclusion (Slee, 2018).

Transactional leadership represents a contrasting orientation based on role clarity, monitoring, exchange relationships, and the achievement of predefined objectives (Bass & Avolio, 1997; Jaqua & Jaqua, 2021). Although often associated with managerial control and organisational stability, transactional leadership may also contribute to inclusion by providing structure, coordination, and clear expectations (Dong, 2023). This distinction may be particularly important for the normative dimensions of school culture. Whereas transformational leadership may influence shared beliefs and expectations by encouraging collective reflection and the reconsideration of established assumptions, transactional leadership is primarily oriented towards regulating behaviour within existing organisational frameworks and may therefore have more limited capacity to reshape deeply embedded normative expectations (Jensen et al., 2019; Lambrecht et al., 2020; Maurice et al., 2024).

Consequently, transformational and transactional leadership should not be viewed as mutually exclusive or inherently inclusive. Transformational leadership may contribute to the cultural and ethical dimensions of inclusion (Carrington et al., 2024), whereas transactional leadership may provide the organisational conditions necessary for implementation (Korkmaz et al., 2022). From a critical perspective, the central issue is not which style is adopted, but whether leadership practices challenge exclusionary assumptions and expand opportunities for participation (Shields, 2026). Inclusive leadership therefore requires more than efficiency or vision alone; it demands a commitment to equity, belonging, and educational justice (Sun et al., 2024; Shah et al., 2022).

2.4. Developing the Conceptual Model

The conceptual model proposed in this study is grounded in the assumption that inclusive school development is shaped by the interplay of leadership behaviours, principal self-efficacy, and school culture rather than any single factor considered in isolation. Leadership influences organisational values and practices, school culture shapes the conditions under which inclusion is supported or resisted, and self-efficacy affects leaders’ willingness and capacity to sustain inclusive action (Carrington et al., 2024; Leithwood & Jantzi, 2006; Li & Liu, 2020; Tsiolpidou & Koutouzis, 2024; Tsirantonaki & Vlachou, 2024b; Zhang et al., 2026). Inclusive practice is therefore conceptualised as the outcome of interconnected organisational and psychological processes.

The first component of the model concerns the relationship between leadership styles and school culture. Transformational leadership is expected to demonstrate stronger positive associations with school culture because it emphasises shared vision, collaboration, and collective commitment (Carrington et al., 2024; Leithwood & Jantzi, 2006). Transactional leadership may also contribute through role clarification, coordination, and organisational stability, although its influence is expected to be comparatively weaker (Bass & Avolio, 1997; Maurice et al., 2024).

The second component focuses on the predictive contribution of principal self-efficacy, leadership styles, and school culture to the implementation of inclusive policies and practices. Based on previous research, self-efficacy is expected to play a particularly important role because it influences leaders’ persistence, initiative, and capacity to manage complex organisational change (Bandura, 1997; Tschannen-Moran & Gareis, 2004).

The third component positions principal self-efficacy as a potential mediating mechanism. Leadership behaviours may influence inclusive practice not only directly but also indirectly by strengthening principals’ beliefs in their capacity to lead change (Verheijen-Tiemstra et al., 2025). In this way, self-efficacy may function as the mechanism through which leadership orientations are translated into inclusive school practices.

On this basis, the following hypotheses were tested:

H1. 

Transformational and transactional leadership are positively associated with overall school culture and its constituent dimensions.

H2. 

Principal self-efficacy, transformational leadership, transactional leadership, and school culture jointly predict the implementation of policies and practices that promote inclusive school culture.

H3. 

Principal self-efficacy mediates the relationships between transformational and transactional leadership and each of the four dimensions of inclusive policy and practice implementation: inclusive leadership and school culture, school–community inclusive policies, inclusive teaching and assessment practices, and student and family empowerment.

3. Method

3.1. Research Design

Thid study formed part of a broader explanatory sequential mixed-methods project examining the relationships among principal self-efficacy, leadership styles, school culture, and inclusive policy and practice implementation in Greek primary and lower-secondary schools (Creswell & Plano Clark, 2011; Schoonenboom & Johnson, 2017). The overall project included a quantitative survey phase followed by a qualitative phase designed to explore and contextualise the quantitative findings.

This article reports exclusively on the quantitative phase. A cross-sectional correlational design was employed using self-report survey data collected from school principals. The design enabled the examination of relationships among leadership styles, principal self-efficacy, school culture, and inclusive leadership practices, as well as the testing of predictive and mediation models through hierarchical regression and mediation analyses.

3.2. Participants

The quantitative phase involved 452 principals serving in public primary and lower-secondary schools across Greece (

Table 1. Demographic and professional characteristics of the study sample (N = 452).

Variable	Category	n	%
Educational Level Served	Primary Education (Elementary School)	322	71.2
	Secondary Education (Lower Secondary School)	130	28.8
Years of Experience as Principal	0–5 years	173	38.3
	6–10 years	66	14.6
	11–15 years	62	13.7
	16–20 years	35	7.7
	More than 20 years	116	25.7
Gender	Female	237	52.4
	Male	212	46.9
	Other	1	0.2
	Prefer not to say	2	0.4
Age	25–35 years	6	1.3
	36–45 years	16	3.5
	46–55 years	125	27.7
	Above 55 years	305	67.5
Highest Educational Qualification	Bachelor’s Degree	87	19.2
	Master’s Degree	305	67.5
	Doctoral Degree (PhD)	39	8.6
	Other	21	4.6
Teaching Experience in General Education	0–4 years	7	1.5
	5–9 years	17	3.8
	10–19 years	26	5.8
	20 years and above	402	88.9
Teaching Experience in Special Education	None or less than 1 year	340	75.2
	1–4 years	47	10.4
	5–9 years	13	2.9
	10–19 years	25	5.5
	20 years and above	27	6.0
School Size	Small (≤100 students)	162	35.8
	Medium (101–250 students)	229	50.7
	Large (251–400 students)	57	12.6
	Very Large (>400 students)	4	0.9
School Location	Island Area (excluding Crete and Euboea)	65	14.4
	Rural Area	138	30.5
	Urban Area	249	55.1

Note. Percentages may not total exactly 100 due to rounding.

). The sample included 322 principals from primary schools (71.2%) and 130 from lower-secondary schools (28.8%).

Participants were predominantly experienced educational leaders. Most were over 55 years of age, held postgraduate qualifications, and reported extensive experience in general education. By contrast, experience in special education was generally limited. Participants represented a wide range of school contexts, including urban, rural, and island schools, as well as schools of varying size.

The sample size was considered adequate for the planned psychometric and multivariate analyses, exceeding commonly recommended requirements for factor analysis, regression, and mediation modelling.

3.3. Measures

Data were collected using a structured questionnaire comprising four validated instruments measuring inclusive leadership practices, school culture, principals’ self-efficacy for inclusive leadership, and leadership style. All instruments were administered in Greek as part of a single online survey.

3.3.1. Leadership for Inclusion Questionnaire (LEI-Q)

Inclusive leadership practices were assessed using the Leadership for Inclusion Questionnaire (LEI-Q) (Crisol-Moya et al., 2020). The instrument was originally developed to evaluate the extent to which school leaders promote inclusive policies, practices, and organisational conditions that support the participation and success of all students. The questionnaire consists of 38 items rated on a four-point Likert-type scale ranging from 1 (Not yet implemented) to 4 (Fully implemented). Higher scores indicate stronger implementation of inclusive leadership practices within the school.

3.3.2. School Culture Survey (SCS)

School culture was measured using the School Culture Survey (SCS) (Higgins-D’Alessandro & Sadh, 1998), which assesses principals’ perceptions of the social and organisational climate of their schools. The instrument contains 25 positively worded items covering normative expectations, student–teacher relationships, student relationships, and educational opportunities. Responses are recorded on a five-point Likert scale ranging from 1 (Strongly disagree) to 5 (Strongly agree), with higher scores reflecting a more positive school culture.

3.3.3. Principals’ Self-Efficacy for Inclusive Leadership Scale

Principals’ perceived self-efficacy for inclusive leadership was assessed using the School Principals’ Self-Efficacy for Inclusive Leadership Scale (Tsiolpidou & Koutouzis, 2024). The instrument evaluates principals’ beliefs regarding their capability to effectively lead inclusive schools, support collaboration, improve teaching and learning processes, and provide strategic direction for inclusive educational development. The scale comprises 20 items rated on a six-point response format ranging from 1 (Cannot do at all) to 6 (Completely certain I can do). Higher scores indicate stronger perceived self-efficacy for inclusive leadership.

3.3.4. Transformational and Transactional Leadership Scale (TATL)

Leadership style was assessed using the Transformational and Transactional Leadership Scale (TATL; Jensen et al., 2019). The instrument measures the extent to which school principals engage in transformational and transactional leadership behaviours. Participants responded to 19 items using a five-point Likert scale ranging from 1 (Strongly disagree) to 5 (Strongly agree). Higher scores indicate greater endorsement of the corresponding leadership style.

3.4. Data Analysis

Data analyses were conducted using IBM SPSS Statistics 27, while confirmatory factor analyses were performed using IBM SPSS AMOS. Prior to hypothesis testing, data were screened for missing values, outliers, and compliance with the assumptions of parametric analyses.

The psychometric properties of all instruments were evaluated before testing the research model. For the Leadership for Inclusion Questionnaire (LEI-Q) and the School Culture Survey (SCS), exploratory and confirmatory factor analyses were conducted to examine their validity within the Greek educational context. For the Principals’ Self-Efficacy for Inclusive Leadership Scale and the Transformational and Transactional Leadership Scale (TATL), confirmatory analyses were conducted based on existing validation evidence. Internal consistency was assessed using Cronbach’s alpha coefficients.

Descriptive statistics were calculated for all study variables. Hypothesis 1 was tested using Pearson product–moment correlation coefficients to examine associations between leadership styles and school culture. Additional bivariate correlations were also computed among principal self-efficacy, leadership styles, school culture, and inclusive leadership practices.

Hypothesis 2 was tested using hierarchical multiple regression analysis, with predictors entered in theoretically informed blocks to assess their incremental contribution to explained variance in inclusive leadership practices. Relevant diagnostics were examined to evaluate model assumptions.

Hypothesis 3 was tested using Hayes’ PROCESS macro for SPSS (Model 4). Separate mediation models were estimated to examine whether principal self-efficacy mediated the relationship between leadership styles and each dimension of inclusive leadership practices. Indirect effects were estimated using 5000 bootstrap resamples and bias-corrected 95% confidence intervals. Statistical significance was set at α = 0.05.

4. Results

4.1. Psychometric Evaluation

Prior to testing the study hypotheses, the psychometric properties of all measurement instruments were evaluated to ensure the reliability and construct validity of the study variables. Because the Leadership for Inclusion Questionnaire (LEI-Q) and the School Culture Survey (SCS) had not previously undergone comprehensive validation within the Greek educational context, both exploratory and confirmatory factor analyses were conducted. In contrast, the Principals’ Self-Efficacy for Inclusive Leadership Scale and the Transformational and Transactional Leadership Scale (TATL) had already demonstrated satisfactory psychometric properties in previous validation studies and were therefore examined through confirmatory procedures and reliability analyses only.

For the LEI-Q, the originally proposed two-factor structure did not adequately capture the multidimensional nature of inclusive leadership within the Greek context. Exploratory factor analysis supported a revised four-factor solution representing: (a) Inclusive Leadership and School Culture, (b) School–Community Inclusive Policies, (c) Inclusive Teaching and Assessment Practices, and (d) Student and Family Empowerment. Confirmatory factor analysis of the revised structure demonstrated good model fit: χ²(431) = 698.49, p < 0.001, χ²/df = 1.62, CFI = 0.94, TLI = 0.93, RMSEA = 0.05, and SRMR = 0.06. The complete cross-validation procedure, including exploratory and confirmatory analyses, factor loadings, reliability estimates, and model evaluation, is reported in the published psychometric validation of the Greek LEI-Q (Tsirantonaki & Vlachou, 2026).

The psychometric evaluation of the SCS supported the retention of the original four-factor structure proposed by the instrument developers. Although exploratory analyses suggested a slightly modified factor configuration, confirmatory analyses indicated that the original model provided superior conceptual clarity and overall model fit. Reliability estimates for all dimensions were high, supporting the use of the original structure in subsequent analyses.

Confirmatory analyses further supported the adequacy of the Principals’ Self-Efficacy for Inclusive Leadership Scale and the TATL instrument. The self-efficacy scale demonstrated excellent internal consistency, indicating that the items measured a highly coherent latent construct. Similarly, both leadership dimensions of the TATL questionnaire exhibited satisfactory reliability, with transformational leadership demonstrating particularly strong internal consistency.

Overall, the psychometric analyses provided strong evidence for the reliability and construct validity of all measures employed in the study (

Table 2. Summary of psychometric properties of the study measures.

Measure	Factor Structure	Internal Consistency (Cronbach’s α)	Evidence of Construct Validity
Leadership for Inclusion Questionnaire (LEI-Q)	Revised four-factor model: (a) Inclusive Leadership and School Culture, (b) School–Community Inclusive Policies, (c) Inclusive Teaching and Assessment Practices, (d) Student and Family Empowerment	0.830–0.941	EFA and CFA supported the revised four-factor structure
School Culture Survey (SCS)	Original four-factor model: (a) Normative Expectations, (b) Student–Teacher Relationships, (c) Student Relationships, (d) Educational Opportunities	0.824–0.942	EFA and CFA supported retention of the original structure
Principals’ Self-Efficacy for Inclusive Leadership Scale	Unidimensional structure	0.969	CFA supported the original theoretical structure
Transformational Leadership (TATL)	Unidimensional higher-order construct	0.935	CFA supported the proposed structure
Transactional Leadership (TATL)	Unidimensional higher-order construct	0.771	CFA supported the proposed structure

Note. Detailed psychometric results for the Greek LEI-Q, including the complete exploratory and confirmatory factor analyses, factor loadings, explained variance, and model evaluation, are reported in the published validation study (Tsirantonaki & Vlachou, 2026).

). While the LEI-Q required a revised four-factor solution to better reflect inclusive leadership practices within the Greek educational context, the remaining instruments retained their original theoretical structures and demonstrated satisfactory model fit and internal consistency. These findings support the suitability of all study measures for the subsequent correlational, regression, and mediation analyses.

4.2. Hypothesis 1: Associations Between Leadership Styles and School Culture

Hypothesis 1 proposed that principals’ leadership styles would be significantly associated with school culture. To test this hypothesis, Pearson product–moment correlations were conducted between transformational leadership, transactional leadership, overall school culture, and the four dimensions of the School Culture Survey (SCS).

As shown in

Table 3. Pearson correlations among the main study variables (N = 452).

Variable	1	2	3	4	5	6	7	8
1. Principal Self-Efficacy	—
2. Transformational Leadership	0.689 ***	—
3. Transactional Leadership	0.449 ***	0.537 ***	—
4. Normative Expectations	0.272 ***	0.199 ***	0.060	—
5. Student–Teacher/School Relationships	0.473 ***	0.355 ***	0.196 ***	0.485 ***	—
6. Student Relationships	0.362 ***	0.266 ***	0.110 *	0.559 ***	0.605 ***	—
7. Educational Opportunities	0.473 ***	0.415 ***	0.261 ***	0.466 ***	0.629 ***	0.671 ***	—
8. Inclusive Leadership Practices	0.671 ***	0.603 ***	0.404 ***	0.172 ***	0.367 ***	0.253 ***	0.382 ***	—

Note. Values are Pearson’s r. N = 452. * p < 0.05, *** p < 0.001. The correlation between Transactional Leadership and Normative Expectations was not statistically significant (r = 0.060, p = 0.203).

, the principal study variables were generally positively correlated. Principal self-efficacy showed strong positive associations with transformational leadership (r = 0.689, p < 0.001) and inclusive leadership practices (r = 0.671, p < 0.001), while transformational leadership was also strongly associated with inclusive leadership practices (r = 0.603, p < 0.001). Correlations among the predictors ranged from weak to strong but remained below levels typically indicative of severe bivariate collinearity. Transformational leadership was positively associated with all four dimensions of school culture, with correlations ranging from r = 0.199 to r = 0.415 (all p < 0.001). Transactional leadership showed weaker associations with school culture and was not significantly associated with Normative Expectations (r = 0.060, p = 0.203), although it was positively associated with Student–Teacher/School Relationships (r = 0.196, p < 0.001), Student Relationships (r = 0.110, p = 0.019), and Educational Opportunities (r = 0.261, p < 0.001).

At the dimensional level, transformational leadership was significantly associated with all four school culture dimensions. The strongest relationship was observed for Educational Opportunities, r(450) = 0.415, p < 0.001, followed by Student–Teacher/School Relationships, r(450) = 0.355, p < 0.001; Student Relationships, r(450) = 0.266, p < 0.001; and Normative Expectations, r(450) = 0.199, p < 0.001. Transactional leadership demonstrated weaker associations, with significant relationships observed for Student–Teacher/School Relationships, r(450) = 0.196, p < 0.001; Student Relationships, r(450) = 0.110, p = 0.019; and Educational Opportunities, r(450) = 0.261, p < 0.001, but not for Normative Expectations.

Overall, the findings support Hypothesis 1 and suggest that both leadership styles were associated with school culture, although transformational leadership demonstrated consistently stronger relationships across all dimensions. These findings further suggest that transformational leadership is more consistently associated with dimensions of school culture that reflect supportive relationships, shared expectations, and enhanced educational opportunities.

4.3. Hypothesis 2: Predicting Inclusive Leadership Practices

Hypothesis 2 proposed that principals’ self-efficacy, leadership styles, and school culture would significantly predict the implementation of inclusive leadership practices. A hierarchical multiple regression analysis was conducted to test this hypothesis, with inclusive leadership practices as the dependent variable. Prior to the analysis, regression assumptions were examined. Diagnostic indicators indicated no evidence of problematic multicollinearity, while residual analyses supported the assumptions of normality, linearity, independence of errors, and homoscedasticity.

The results of the hierarchical multiple regression analysis are presented in

Table 4. Hierarchical multiple regression predicting inclusive leadership practices (N = 452).

Step/Predictor	B	SE	β	p
Step 1
Principal Self-Efficacy	0.45	0.02	0.67	<0.001
Step 2
Principal Self-Efficacy	0.32	0.03	0.49	<0.001
Transformational Leadership	0.24	0.04	0.27	<0.001
Step 3
Principal Self-Efficacy	0.30	0.03	0.45	<0.001
Transformational Leadership	0.21	0.04	0.23	<0.001
Transactional Leadership	0.06	0.04	0.05	n.s.
Normative Expectations	−0.03	0.03	−0.04	n.s.
Student–Teacher/School Relationships	0.06	0.05	0.06	n.s.
Student Relationships	−0.02	0.04	−0.03	n.s.
Educational Opportunities	0.05	0.04	0.06	n.s.

Note. B = unstandardised regression coefficient; SE = standard error; β = standardised regression coefficient. Step 1: R² = 0.45. Step 2: R² = 0.49, ΔR² = 0.04. Step 3: R² = 0.50, ΔR² = 0.01. n.s. = not statistically significant.

. The model was estimated in three steps. In the first step, principals’ self-efficacy emerged as a strong predictor, accounting for 45% of the variance in inclusive leadership practices (R² = 0.45, β = 0.67, p < 0.001). The addition of transformational leadership in the second step significantly improved the model, increasing the explained variance by an additional 4% (ΔR² = 0.04). Transformational leadership remained a significant predictor at this step (β = 0.27, p < 0.001). In the final step, the inclusion of transactional leadership and the four school culture dimensions produced only a marginal improvement in model fit, increasing the explained variance by an additional 1% (ΔR² = 0.01). The final model accounted for 50% of the variance in inclusive leadership practices (R² = 0.50).

In the final model, self-efficacy (β = 0.45, p < 0.001) and transformational leadership (β = 0.23, p < 0.001) remained significant predictors, whereas transactional leadership and the school culture dimensions did not demonstrate unique predictive effects. These findings support Hypothesis 2 and highlight principals’ self-efficacy as the strongest predictor of inclusive leadership practices, with transformational leadership providing an additional but smaller contribution.

4.4. Hypothesis 3: The Mediating Role of Principals’ Self-Efficacy

To examine Hypothesis 3, eight separate mediation models were estimated using PROCESS Model 4 with 5000 bootstrap resamples (N = 452). Transformational and transactional leadership served separately as predictor variables, principals’ self-efficacy as the mediator, and each of the four dimensions of inclusive leadership practices as the respective outcome variable (Figure 1).

Results consistently demonstrated significant indirect effects of both leadership styles on all four dimensions of inclusive leadership practices through principals’ self-efficacy. Statistical significance was determined using bias-corrected 95% bootstrap confidence intervals, with mediation considered significant when the confidence interval did not include zero.

For transformational leadership, significant indirect effects were observed for Inclusive Leadership and School Culture (Effect = 0.256, 95% CI [0.171, 0.343]), School–Community Inclusive Policies (Effect = 0.308, 95% CI [0.237, 0.386]), Inclusive Teaching and Assessment Practices (Effect = 0.357, 95% CI [0.266, 0.450]), and Student and Family Empowerment (Effect = 0.289, 95% CI [0.197, 0.376]).

Similarly, significant indirect effects were identified for transactional leadership across all four dimensions: Inclusive Leadership and School Culture (Effect = 0.271, 95% CI [0.200, 0.347]), School–Community Inclusive Policies (Effect = 0.304, 95% CI [0.231, 0.386]), Inclusive Teaching and Assessment Practices (Effect = 0.294, 95% CI [0.221, 0.374]), and Student and Family Empowerment (Effect = 0.277, 95% CI [0.171, 0.392]) (

Table 5. Direct and indirect effects of leadership styles on inclusive leadership practices through principals’ self-efficacy (N = 452).

Leadership Style	Outcome Dimension	Direct Effect (c′)	Indirect Effect (ab)	95% Bootstrap CI
Transformational	Inclusive Leadership & School Culture	0.273	0.256	[0.171, 0.343]
Transformational	School–Community Inclusive Policies	0.244	0.308	[0.237, 0.386]
Transformational	Inclusive Teaching & Assessment Practices	0.202	0.357	[0.266, 0.450]
Transformational	Student & Family Empowerment	0.188	0.289	[0.197, 0.376]
Transactional	Inclusive Leadership & School Culture	0.122	0.271	[0.200, 0.347]
Transactional	School–Community Inclusive Policies	0.104	0.304	[0.231, 0.386]
Transactional	Inclusive Teaching & Assessment Practices	0.267	0.294	[0.221, 0.374]
Transactional	Student & Family Empowerment	0.062	0.277	[0.171, 0.392]

Note. N = 452. Mediator = Principals’ Self-Efficacy. Direct effect (c′) represents the association between leadership style and the respective outcome after accounting for principals’ self-efficacy. Indirect effects (ab) were estimated using PROCESS Model 4 with 5000 bootstrap resamples. An indirect effect was considered statistically significant when the 95% bootstrap confidence interval did not include zero.

).

Across all analyses, none of the bootstrap confidence intervals included zero, indicating that principals’ self-efficacy functioned as a significant mediator between leadership styles and the implementation of inclusive leadership practices. These findings support Hypothesis 3 and indicate significant indirect associations between leadership styles and inclusive school development through principals’ self-efficacy. Given the cross-sectional design, these indirect effects should not be interpreted as evidence of temporal or causal ordering.

5. Discussion

This study examined how principals’ leadership styles, perceived self-efficacy, and school culture are associated with the implementation of inclusive policies and practices. Three findings organise the discussion. First, both transformational and transactional leadership were positively associated with school culture, although transformational leadership displayed consistently stronger relationships. Second, principals’ self-efficacy emerged as the dominant predictor of reported inclusive leadership practices, while transformational leadership made a smaller but meaningful incremental contribution. Third, the mediation analyses produced significant indirect effects through self-efficacy for both leadership styles and across all four dimensions of inclusive practice. Taken together, these findings challenge accounts of inclusion that reduce school improvement to the adoption of a preferred leadership style. They suggest instead that leadership becomes educationally consequential when it is translated into an agentic belief that inclusive change is both possible and actionable (Bandura, 1997; Carrington et al., 2024; Tsiolpidou & Koutouzis, 2024; T. Wang et al., 2022).

5.1. Leadership and School Culture: Transformative Potential and Structural Limits

The positive association between leadership styles and school culture reinforces the proposition that culture is not merely the background against which leadership occurs. It is one of the principal terrains upon which leadership is exercised. Principals communicate what is valued, whose participation matters, which forms of achievement are recognised, and which difficulties are treated as organisational responsibilities rather than individual deficiencies. Through these processes, leadership contributes to the construction of the school’s normative and relational order (Bush, 2020; Day et al., 2020; DeMatthews et al., 2020; Gu, 2023; Leithwood et al., 2020).

Transformational leadership was more strongly related to overall school culture than transactional leadership. Its strongest association concerned educational opportunities, followed by student–teacher and school relationships. This pattern is theoretically significant. Transformational leadership appears to be connected most clearly with those dimensions of culture through which schools distribute participation, support learning, and construct relational belonging (Carrington et al., 2024; Wilson Heenan et al., 2023). A shared vision, however, matters only insofar as it changes what students and educators are permitted, expected, and supported to do. Otherwise, the language of vision may remain largely rhetorical, appearing prominently in mission statements while having limited influence on the institutional routines through which some students continue to be positioned at the margins (Graham, 2020; Slee, 2018).

The findings therefore offer qualified support for transformational leadership. They are consistent with research linking transformational practices to collective purpose, professional commitment, and positive organisational conditions (Leithwood & Jantzi, 2006; Leithwood & Sun, 2012). Yet the correlations were moderate rather than overwhelming. Transformational leadership should not be regarded as a universal solution capable of overcoming entrenched structural inequalities. A principal may articulate inclusion while governing within systems structured by standardisation, competitive accountability, insufficient specialist provision, and historically sedimented assumptions about ability. Leadership can contest these arrangements, but it does not stand outside them (Graham, 2020; Tsirantonaki & Vlachou, 2024b; Vlachou, 2023).

The relatively weak relationship between transformational leadership and normative expectations is revealing in this regard. Normative expectations are often embedded in established disciplinary traditions, achievement hierarchies, and institutionally sanctioned ideas of appropriate behaviour. Such arrangements may prove more resistant to visionary influence than relational or opportunity-oriented aspects of culture (Slee, 2018; Spandagou, 2021). This lends support to critical arguments that exclusion is not simply produced by negative attitudes that can be corrected through inspirational leadership. It is institutionalised through ordinary rules, classifications, routines, and judgements that present themselves as neutral (Slee, 2018).

Transactional leadership showed weaker but still significant associations with overall school culture and three of its four dimensions. It would therefore be simplistic to dismiss transactional leadership as inherently incompatible with inclusion (Lambrecht et al., 2020; Maurice et al., 2024). Schools require coordination, clarity, feedback, accountability, and reliable organisational procedures. Inclusive intentions cannot be sustained by moral aspiration alone. They require timetables, resource allocation, monitoring, role definition, and follow-through. Transactional practices may thus provide part of the organisational infrastructure through which inclusive commitments are implemented (Dong, 2023; Korkmaz et al., 2022).

Their limits, however, are equally apparent. Systems of reward, monitoring, and compliance can secure procedural order without disturbing the assumptions that produce exclusion. A school may comply with formal inclusion requirements while retaining deficit-based expectations, segregating responsibilities, and restricting meaningful participation (Graham, 2020; Slee, 2018; Vlachou, 2023). The absence of a significant relationship between transactional leadership and normative expectations is particularly instructive. Leadership focused on rules does not necessarily transform the normative foundations from which those rules derive. Compliance can reproduce the prevailing order as efficiently as it implements reform.

5.2. Self-Efficacy and the Capacity to Enact Inclusive Practice

The hierarchical regression analysis identifies principals’ self-efficacy as the strongest explanatory factor in the empirical model. When entered alone, self-efficacy explained 45% of the variance in inclusive leadership practices. Transformational leadership added a further 4%, whereas transactional leadership and the dimensions of school culture collectively added only 1% in the final step. In that final model, self-efficacy and transformational leadership remained significant, while transactional leadership and the school-culture dimensions did not retain unique predictive effects.

This does not mean that culture and transactional leadership are irrelevant. Rather, their shared explanatory variance appears to overlap substantially with self-efficacy and transformational leadership. The result also cautions against treating statistically significant bivariate relationships as evidence of independent explanatory power. School culture may be closely related to leadership but may not operate as a separate predictor once the principals’ perceived capacity to act and their transformational behaviours are considered simultaneously (Liu, 2024; T. Wang et al., 2022).

The prominence of self-efficacy is consistent with Bandura’s (1997, 2012) account of efficacy beliefs as regulators of effort, persistence, strategic adaptation, and resilience. Inclusive leadership requires principals to act in conditions where goals are contested, resources are uneven, and outcomes are rarely immediate. It involves challenging established expectations, coordinating professional collaboration, supporting differentiated teaching, engaging families, and sustaining reform in the face of resistance. Under such conditions, knowledge of inclusive principles is insufficient. Principals must also believe that they can mobilise the actions required to enact them (Tschannen-Moran & Gareis, 2004; Tsiolpidou & Koutouzis, 2024; Van Staden-Payne & Nel, 2023; T. Wang et al., 2022).

The finding thus helps explain the recurrent gap between endorsement and implementation. Inclusion is widely affirmed in policy discourse; its principles are now difficult to oppose publicly. Yet rhetorical commitment can coexist comfortably with practices that continue to marginalise. Self-efficacy may constitute one of the mechanisms distinguishing symbolic agreement from sustained action. Principals who perceive themselves as capable of coordinating change may be more willing to persist when inclusive reforms disrupt established routines or generate professional and parental contestation (Schmitz & Tulowitzki, 2025; Röhl et al., 2024).

Nevertheless, this finding must not be converted into another individualising account of school failure. The language of self-efficacy can easily be appropriated to imply that insufficient inclusion reflects insufficient confidence among principals. Such an interpretation would reproduce precisely the deficit logic that inclusive education seeks to contest. Principals do not construct efficacy beliefs in a vacuum. These beliefs are shaped by professional preparation, prior success, institutional support, access to expertise, organisational autonomy, and the availability of material and human resources, while accountability and performativity pressures may simultaneously delimit the forms of action that principals perceive as institutionally possible or legitimate. Asking principals simply to “believe more strongly” while leaving exclusionary structures untouched would replace reform with motivational rhetoric (Hamann et al., 2024; Mustafa et al., 2019; Slee, 2011).

Self-efficacy should therefore be interpreted as relationally and institutionally produced (Hamann et al., 2024; Mustafa et al., 2019). It represents neither an innate leadership trait nor a substitute for structural support. High self-efficacy may enable action, but confidence without knowledge, resources, or critical reflexivity can become misplaced certainty. Equally, low self-efficacy may be a rational response to systems that allocate responsibility for inclusion without providing the means to enact it. The practical implication is not merely to strengthen principals psychologically, but to create conditions in which inclusive action can generate credible experiences of professional mastery (Bandura, 1997; Federici & Skaalvik, 2011).

The relatively modest incremental contribution of transformational leadership is also important. Although transformational behaviours remained significant, much of their association with inclusive practice overlapped with self-efficacy. Visionary leadership may matter partly because it is accompanied by a stronger sense of agency. A vision becomes operational when principals believe they can convene others around it, withstand resistance, and reorganise practice. Without that capacity, transformational leadership risks becoming a language of aspiration detached from implementation (Gümüş et al., 2026; Z. Wang et al., 2026; Zhang et al., 2026).

The loss of statistical significance of the school culture dimensions in the final regression model should be interpreted in a similar way. Their significant bivariate associations with inclusive leadership practices indicate that school culture remains relevant; however, once principal self-efficacy and transformational leadership were considered simultaneously, the culture dimensions did not explain substantial unique variance beyond these stronger predictors. This attenuation is likely to reflect shared explanatory variance among conceptually related organisational and leadership constructs rather than an absence of association between school culture and inclusive practice. Importantly, the regression diagnostics did not indicate problematic multicollinearity. The findings therefore suggest that leadership, self-efficacy, and school culture operate as interconnected dimensions of school organisation whose effects are difficult to isolate statistically when modelled concurrently.

5.3. Self-Efficacy as an Indirect Pathway Between Leadership and Inclusion

The mediation analyses deepen this interpretation. Significant indirect effects through principals’ self-efficacy were identified for both transformational and transactional leadership across all four dimensions examined in the model: Inclusive Leadership and School Culture, School–Community Inclusive Policies, Inclusive Teaching and Assessment Practices, and Student and Family Empowerment. The consistency of these indirect associations is one of the study’s most important contributions.

The results suggest that leadership style alone does not fully account for inclusive practice. Leadership behaviours appear to be associated with the extent to which principals perceive themselves as capable of organising and sustaining inclusive action, and this perceived capability is, in turn, associated with implementation. In this sense, self-efficacy may be understood as a translation point between leadership orientation and organisational practice (Verheijen-Tiemstra et al., 2025; T. Wang et al., 2022).

For transformational leadership, the indirect effects are theoretically coherent. Developing and communicating a collective vision, encouraging professional reflection, and mobilising shared goals may reinforce principals’ confidence in their capacity to lead change (Carrington et al., 2024; Leithwood & Jantzi, 2006; Z. Wang et al., 2026). That perceived capacity may then support action across policy, pedagogy, school–community collaboration, and student and family participation. The largest indirect effect was observed for Inclusive Teaching and Assessment Practices, suggesting that self-efficacy may be especially consequential where leadership must move beyond broad values and influence technically and politically demanding areas of instructional practice (Tsiolpidou & Koutouzis, 2024; T. Wang et al., 2022).

The significant indirect effects found for transactional leadership complicate the common contrast between transformational leadership as progressive and transactional leadership as merely bureaucratic (Lambrecht et al., 2020; Maurice et al., 2024). Transactional practices may contribute to self-efficacy when they provide role clarity, feedback, predictable procedures, recognition, and organisational control. In contexts of uncertainty, these features may help principals perceive inclusive initiatives as manageable rather than abstract. Transactional leadership may therefore support implementation indirectly, even if it is less strongly connected with the deeper relational and cultural dimensions of schooling (Dong, 2023; Korkmaz et al., 2022).

This should not be read as evidence that any form of managerial control becomes inclusive when accompanied by confidence. Transactional leadership can facilitate implementation, but it can also narrow inclusion into compliance with procedures. Its educational meaning depends on what is being monitored, rewarded, and normalised. A technically efficient school may still be profoundly exclusionary (Graham, 2020; Slee, 2018). The key question is not whether organisational routines exist, but whether they redistribute participation and challenge the institutional production of marginality (Sun et al., 2024; Shields, 2026).

More broadly, the mediation findings shift the analytical question from “Which leadership style works best?” to “Through what conditions do leadership behaviours become actionable?” This is a more productive question because it resists the search for universally effective leadership formulas. Inclusive leadership is not a package of behaviours that can be transferred unchanged across schools (DeMatthews et al., 2020; Tsirantonaki & Vlachou, 2024b). It is situated work undertaken within specific cultural, political, and material conditions. Self-efficacy appears to be one important part of that process, but it is not the process in its entirety (Graham, 2020; Hardy & Woodcock, 2024; Vlachou, 2023).

The cross-sectional design also requires caution. Although the proposed models are theoretically grounded, temporal ordering cannot be established from the present data. Successful implementation may strengthen principals’ self-efficacy, just as self-efficacy may support implementation (Bandura, 1997, 2012). Similarly, positive school conditions may reinforce particular leadership behaviours. The mediation findings should therefore be understood as evidence of statistically significant indirect associations consistent with the conceptual model, rather than as proof of a unidirectional causal sequence.

5.4. Reframing Inclusive Leadership: From Individual Agency to Systemic Conditions

Across the three hypotheses, a coherent picture emerges. Transformational and transactional leadership are associated with school culture, but transformational leadership has the stronger and more consistent relationship. Self-efficacy emerged as the strongest explanatory factor in the model of inclusive practice, while transformational leadership retains an additional independent contribution. Self-efficacy also constitutes a significant indirect pathway linking both leadership styles to all four dimensions of inclusive practice (Verheijen-Tiemstra et al., 2025; T. Wang et al., 2022).

The study’s central contribution is therefore not the claim that confident principals solve exclusion. It is the identification of self-efficacy as a pivotal point at which leadership orientations may be converted into action. This distinction matters. Inclusion has too often been rendered as a technical programme to be delivered by sufficiently committed individuals. Such an account leaves the architecture of exclusion largely intact and risks attributing the failure of reforms to materialise primarily to practitioners rather than to the structural conditions that shape inclusive practice (Graham, 2020; Slee, 2011, 2018).

A more defensible interpretation is that inclusive leadership requires both agentic principals and enabling institutions. Professional learning must develop not only procedural knowledge but also critical capacity: the ability to identify how apparently neutral practices classify, separate, and diminish students (Slee, 2018; Spandagou, 2021; Vlachou, 2023). Organisational systems must provide resources, collaboration, specialist support, and meaningful room for professional judgement (Bakogiannis & Papavasiliou, 2026; Hardy & Woodcock, 2024; Tsirantonaki & Vlachou, 2024b). Accountability must examine participation, belonging, and equitable opportunity rather than merely procedural compliance or aggregate attainment.

Self-efficacy is consequently best viewed as necessary but insufficient. It may help principals persist, coordinate action, and confront resistance (Bandura, 1997, 2012; Tsiolpidou & Koutouzis, 2024). It cannot, by itself, redistribute resources, redesign national accountability regimes, or dismantle entrenched hierarchies of ability and worth (Graham, 2020; Slee, 2011; Vlachou, 2023). Inclusive school transformation will remain fragile when systems celebrate leadership agency while withholding the conditions that make such agency effective. The task is not to manufacture heroic principals, but to build schools and systems in which inclusive action is collectively possible, professionally supported, and institutionally sustained (Ainscow, 2020; Carrington, 2022; DeMatthews et al., 2020).

6. Practical Implications

The findings have important implications for leadership preparation and professional development in inclusive education. While leadership styles were associated with school culture and inclusive practices, principals’ self-efficacy emerged as the strongest predictor of inclusive leadership practices within the model. Consequently, professional learning initiatives should move beyond the transmission of knowledge about inclusion and focus on strengthening principals’ confidence and capacity to lead inclusive change in complex school environments (Bandura, 1997; Röhl et al., 2024; T. Wang et al., 2022).

Leadership development programmes should provide opportunities for authentic problem-solving, collaborative reflection, mentoring, and engagement with real inclusion-related challenges, while also encouraging principals to critically examine how institutional routines, organisational practices, and policy expectations may inadvertently reproduce exclusion or constrain participation. Such experiences may enhance principals’ perceptions of efficacy by creating opportunities for mastery experiences, professional feedback, collective learning, and critical reflection (Bandura, 1997, 2012; Federici & Skaalvik, 2011). At the policy level, support structures should not only promote inclusive values but also ensure that school leaders have access to the resources, expertise, and organisational conditions required to enact them (Graham, 2020; Tsirantonaki & Vlachou, 2024b; Vlachou, 2023). Strengthening inclusive leadership therefore requires attention to both individual agency and the systemic conditions that enable inclusive practice (DeMatthews et al., 2020; Slee, 2011).

7. Limitations

Several limitations should be considered when interpreting the findings. First, the study employed a cross-sectional design, which does not permit causal inferences regarding the relationships among leadership styles, self-efficacy, school culture, and inclusive practices. Although the mediation analyses were theoretically grounded, the directionality of these relationships cannot be conclusively established. Furthermore, the present article reports only the quantitative phase of the broader mixed-methods study and is therefore limited to patterns captured through self-report survey data. Such data cannot fully illuminate the micropolitical processes, informal resistance, negotiated relationships, and context-specific organisational dynamics through which inclusive policies are enacted, contested, or reshaped within schools. The qualitative phase of the broader study is intended to examine these processes in greater depth.

Second, all primary study variables were collected simultaneously through principals’ self-reports, creating the possibility of common method variance and socially desirable responding. This limitation is particularly relevant in inclusion research, where principals may experience normative pressure to report stronger inclusive implementation and greater confidence in their leadership capacity. Future studies could therefore incorporate multiple sources of evidence, including teachers, students, and parents, as well as observational data, to provide a more comprehensive assessment of inclusive leadership practices. Third, the study was conducted within the Greek educational context; therefore, caution is required when generalising the findings to other educational systems with different institutional and policy environments. Longitudinal and comparative studies would provide valuable insights into how leadership, self-efficacy, and inclusion interact across contexts and over time (Lambrecht et al., 2020). In addition, the analyses did not model potential clustering associated with broader geographical or administrative contexts; consequently, unobserved regional or local organisational influences cannot be ruled out. Future studies with explicit regional identifiers and sufficiently distributed clusters could examine such contextual variation through multilevel modelling.

8. Conclusions

This study advances understanding of inclusive educational leadership by identifying self-efficacy as a key indirect pathway linking leadership orientations with inclusive practice. Although transformational leadership displayed stronger associations with school culture and inclusive practices than transactional leadership, principals’ self-efficacy emerged as the most influential predictor and a significant mediating mechanism linking leadership styles to inclusive action (Verheijen-Tiemstra et al., 2025; T. Wang et al., 2022).

The findings suggest that inclusive leadership cannot be understood solely in terms of leadership behaviours or organisational characteristics. Rather, it appears to depend on principals’ perceived capacity to mobilise people, resources, and practices in support of inclusion (Bandura, 1997; Federici & Skaalvik, 2011; Tschannen-Moran & Gareis, 2004). At the same time, self-efficacy should not be interpreted as a substitute for structural and institutional support. Inclusive education requires both capable school leaders and educational systems that provide the conditions necessary for inclusive practice to flourish (Graham, 2020; Slee, 2011; Vlachou, 2023). Future research should continue to examine how individual agency and systemic factors interact in shaping inclusive school transformation (Ainscow, 2020; DeMatthews et al., 2020; Mavrou et al., 2025).