Abstract
This study conceptualises principal well-being as an organisational and relational phenomenon by integrating relational trust into the job demands–resources (JD–R) framework. Responding to calls in educational leadership research to move beyond individualised resilience perspectives, the study explores how relational conditions shape the experience and sustainability of leadership work. A sequential explanatory mixed-methods design was employed, combining structural equation modelling of survey data from 232 primary school principals in Cyprus with qualitative case studies involving interviews, reflective journals, and document analysis. Quantitative findings indicate that job demands are negatively associated with principal well-being, while job resources and relational trust exert significant positive effects. Relational trust moderates the relationship between job demands and well-being, attenuating the negative association under conditions of high trust. In addition, principal well-being functions as a mediating mechanism linking organisational conditions to perceived school climate and teacher trust. Qualitative findings illuminate how trust-based relational practices enable shared responsibility, emotional containment, and collective sense-making, thereby shaping how leadership demands are interpreted and managed in practice. Situated within a highly centralised education system, the study offers a contextually grounded extension of the JD–R framework by positioning relational trust as a meta-resource and conceptualising principal well-being as a relational ecology embedded in organisational routines.
Keywords
Introduction
Educational systems internationally are facing growing concerns regarding the sustainability of school leadership. Over the past two decades, the principalship has expanded into a role characterised by intensified accountability (McKay, 2018), administrative overload (Tintoré et al., 2022), coping with digital transformation and the use of AI (Lipsou et al., 2026), post-pandemic recovery (Chatzipanagiotou and Katsarou, 2023) and increasing expectations for inclusive practice, instructional quality, and staff well-being (Ertem, 2024; Leithwood et al., 2020). These pressures have coincided with consistently high levels of stress, emotional exhaustion, and attrition among school leaders, raising important questions about the long-term sustainability of leadership capacity within schools.
Within educational management scholarship, principal well-being is no longer viewed as a peripheral or purely personal concern. A growing body of research demonstrates that leaders’ psychological and emotional functioning is closely linked to organisational trust, school climate, teacher retention, and the sustainability of improvement efforts (Chen et al., 2025; Collie, 2022; Doyle Fosco, 2022). Principals experiencing chronic strain may struggle to sustain relational leadership practices, distribute responsibility, or maintain coherence during periods of change. Conversely, leaders who maintain well-being are more likely to foster collaborative cultures and trust-based organisational conditions. Despite this recognition, conceptualisations of principal well-being remain theoretically and empirically fragmented and frequently individualised, often framed through resilience (Wang, 2024) or coping lenses that underplay the organisational and relational conditions shaping leaders’ everyday experiences.
These concerns become especially acute in education systems where school leadership is exercised within tightly centralised governance arrangements and limited school-level discretion. In Cyprus, public primary school principals operate in a system where key areas of staffing, funding, and broader school decision-making remain largely controlled at central level, while school leaders continue to carry substantial responsibility for policy implementation, organisational coordination, and everyday problem-solving (Karamanidou, 2025; Keravnos et al., 2025; Pashiardis and Kafa, 2022). Under such conditions, principal well-being is not simply a matter of individual welfare but a significant organisational and system-level issue, because leadership strain may directly affect schools’ capacity to function effectively under conditions of sustained external pressure.
The job demands–resources (JD–R) model offers a well-established framework for understanding occupational well-being by conceptualising it as a dynamic balance between JD–R (Bakker and Demerouti, 2017; Demerouti et al., 2001). Applied to educational leadership, the model has demonstrated strong explanatory value, with excessive demands consistently associated with burnout and disengagement, and organisational resources such as autonomy and collegial support linked to higher engagement and job satisfaction (Chen et al., 2025; Kaufman et al., 2022). However, two limitations are evident in how the JD–R framework has been applied within school leadership research. First, demands and resources are often treated as relatively objective features of work, with limited attention to the relational processes through which they are interpreted and experienced. Second, the framework has largely been operationalised at the individual level, under-theorising the organisational and relational conditions that shape how leadership pressure is absorbed, shared, or amplified (Granziera et al., 2021; Tummers and Bakker, 2021).
Given the inherently relational and emotionally embedded nature of school leadership, these limitations are non-trivial. Leadership work unfolds through daily interactions with teachers, students, parents, and system actors, rendering relational dynamics central to how demands are experienced and resources are mobilised. Relational trust, in particular, offers a theoretically powerful yet underutilised lens for extending JD–R applications in educational management. Trust in schools has been conceptualised as shared confidence in the benevolence, reliability, integrity, competence, fairness, and openness of organisational members (Bryk and Schneider, 2002; Hoy and Tschannen-Moran, 2003; Keravnos and Symeou, 2024). Research also links relational trust in schools to collaboration, professional learning, and positive organisational climates (Bryk and Schneider, 2002; Collie, 2022; Leis et al., 2017). However, trust has rarely been examined as a condition shaping principals’ own well-being, and even less frequently as a mechanism that alters how leadership demands are subjectively experienced.
Emerging organisational scholarship suggests that trust may function as a meta-resource within the JD–R framework, conditioning the experience and mobilisation of both demands and resources (Cann et al., 2022; Hobfoll, 2011). In trust-rich environments, principals may be better positioned to share responsibility, seek support, and engage in collective sense-making without fear of negative evaluation. In low-trust contexts, similar structural conditions may be experienced as isolating, evaluative, and emotionally depleting. From this perspective, leadership strain is not solely a function of workload intensity, but of the relational infrastructure through which demands are encountered.
This study responds to these conceptual gaps by integrating relational trust into the JD–R framework to reconceptualise principal well-being as a relational and organisational phenomenon rather than an individual psychological attribute. Using a sequential explanatory mixed-methods design, the study examines how job demands, job resources, and relational trust interact to shape principal well-being, and how well-being, in turn, functions as a mechanism linking organisational conditions to broader school outcomes. In doing so, the study advances a relational ecology of well-being, conceptualising leadership sustainability as co-produced through structural arrangements, trust-based relationships, and everyday organisational practices.
The study is situated within Cypriot primary education, a context that is especially informative for examining these dynamics because structural constraint and relational proximity coexist in distinctive ways. Public schools operate within a strongly centralised administrative system in which autonomy remains restricted, yet principals remain responsible for managing the daily organisational and relational complexity of school life (Pashiardis et al., 2018). At the same time, the relatively small scale of schools and the close proximity among teachers, families, and community actors render interpersonal relationships especially consequential in the enactment of leadership. In such settings, relational trust is likely to matter not only as a desirable cultural feature, but as a practical condition shaping whether demands are experienced as collectively manageable or individually burdensome (Keravnos and Symeou, 2024). Cyprus therefore provides a theoretically informative setting for examining how relational trust interacts with structural demands to shape principal well-being under pressure, while also offering contextually grounded insight into leadership sustainability in a centralised education system.
Guided by the JD–R framework (Bakker and Demerouti, 2017) and theories of relational trust (Bryk and Schneider, 2002; Hoy and Tschannen-Moran, 2003), the study addresses the following research questions:
What levels of well-being do Cypriot primary school principals report, and how are these associated with perceived JD–R? To what extent does relational trust moderate the relationship between job demands and principal well-being, and how is this reflected in principals’ lived experiences? How do principals describe the organisational and relational conditions that sustain or undermine their well-being, and how do these processes relate to broader school-level outcomes such as school climate and teacher trust?
By addressing these questions, the study makes three contributions to educational management scholarship. First, it extends the JD–R framework by empirically positioning relational trust as a moderating and mediating organisational condition. Second, it reframes principal well-being as a relational and organisational construct embedded in everyday leadership practice rather than an individual attribute. Third, it advances the concept of a relational ecology of well-being as an analytical lens for understanding sustainable school leadership in demanding organisational contexts. In addition, the study offers contextually grounded insight for Cyprus by showing how leadership sustainability in a highly centralised system depends not only on structural resources, but also on the relational conditions through which pressure is interpreted, shared, and managed in everyday school life.
Conceptual framework
Educational leadership is increasingly conceptualised as an organisational, emotional, and moral practice rather than a purely technical or managerial function. Within this perspective, principal well-being constitutes a strategic organisational condition that shapes leaders’ capacity to sustain relational work, distribute responsibility, and maintain coherence during periods of change. Rather than treating well-being as a private psychological outcome, this study conceptualises it as a multidimensional and relationally embedded condition encompassing psychological vitality, emotional regulation, relational connectedness, and sustained professional purpose.
The job demands–resources model in educational leadership
The JD–R model conceptualises occupational well-being as the dynamic balance between JD–R (Bakker and Demerouti, 2017; Demerouti et al., 2001). Job demands refer to aspects of work that require sustained physical or psychological effort, such as workload intensity, emotional labour, and accountability pressure, while job resources encompass factors that support goal attainment, buffer demands, and facilitate professional growth, including autonomy, collegial support, and access to professional learning.
Applied to educational leadership, the JD–R model has demonstrated strong explanatory value. Empirical research consistently shows that excessive demands are associated with burnout, disengagement, and attrition among principals, while organisational resources are linked to higher engagement, satisfaction, and leadership retention (Chen et al., 2025; Kaufman et al., 2022). However, applications of the JD–R framework in educational management research have tended to operationalise demands and resources as relatively objective or static features of work, paying limited attention to the relational processes through which they are interpreted, enacted, and sustained (Granziera et al., 2021; Tummers and Bakker, 2021).
This limitation is particularly salient in school leadership contexts, where demands and resources are experienced through dense interpersonal relationships and emotionally charged organisational environments. Leadership work unfolds through ongoing interaction with teachers, students, parents, and system actors, rendering the subjective experience of pressure inseparable from organisational trust, psychological safety, and relational norms. Accordingly, extending the JD–R framework to account more explicitly for relational conditions is essential for understanding leadership well-being in educational settings.
Relational trust as a meta-resource
Relational trust provides a theoretically robust lens for addressing this gap. In school contexts, trust has been conceptualised as shared confidence in the benevolence, reliability, integrity, competence, fairness, and openness of organisational members (Hoy and Tschannen-Moran, 2003; Keravnos and Symeou, 2024; Tschannen-Moran, 2014). Relational trust specifically refers to the quality of interpersonal relationships among school actors, including principals, teachers, students, and parents, and is enacted through everyday interactions within the school community (Bryk and Schneider, 2002). Unlike broader constructs such as organisational trust or institutional trust, which refer to confidence in formal systems or organisational structures, relational trust captures the interpersonal dynamics through which cooperation, openness, and shared responsibility develop in school settings (Niedlich et al., 2021).
A growing body of empirical research in educational settings has linked relational trust to collaboration, distributed leadership, professional learning, and positive organisational climates within schools (Bryk and Schneider, 2002; Collie, 2022; Leis et al., 2017). However, existing studies have focused primarily on how relational trust shapes teacher collaboration, school improvement processes, and student outcomes, while comparatively little attention has been paid to how relational trust influences principals’ own occupational well-being. This perspective aligns with research on caring and relational leadership, which emphasises relational care and trust as organisational conditions through which leadership capacity is sustained under pressure (Louis et al., 2016).
In contrast to studies that treat trust primarily as an outcome of leadership practice or as a contextual variable, this study conceptualises relational trust as a meta-resource that conditions the experience and mobilisation of both demands and resources. Drawing on Conservation of Resources theory (Hobfoll, 2011), trust is understood as enabling resource gain and buffering resource loss by shaping how individuals interpret and respond to pressure. Within the JD–R framework, organisational resources are expected to buffer the negative effects of demands; however, the accessibility and mobilisation of those resources often depend on relational conditions such as psychological safety, openness, and trust among organisational members.
In trust-rich environments, principals are more likely to delegate leadership, disclose uncertainty, and engage in collective sense-making without fear of negative evaluation (Keravnos et al., 2025; Spillane, 2005). Under such conditions, leadership demands are more likely to be experienced as shared, meaningful, and manageable rather than isolating and personally burdensome.
Conversely, in low-trust contexts, similar structural demands may be experienced as evaluative, risky, and emotionally depleting (Costa et al., 2018). Limited psychological safety constrains help-seeking and relational openness, intensifying the subjective burden of leadership work. Research on principal loneliness and leadership isolation suggests that the absence of trust frequently manifests not as reduced collaboration alone, but as emotional withdrawal and heightened vulnerability, even in otherwise well-resourced organisational settings (Dor-Haim and Oplatka, 2021). Evidence from high-trust primary schools in Cyprus further indicates that trust-based relational practices stabilise leadership functioning by enabling shared responsibility and emotional containment (Keravnos and Symeou, 2024). From this perspective, trust does not merely enhance leadership effectiveness; it shapes the relational conditions under which leadership work remains psychologically sustainable.
From individual resilience to a relational ecology of well-being
Prevailing approaches to principal well-being have often emphasised individual resilience, coping strategies, and personal stress management. While such approaches offer partial insight, they risk individualising responsibility for strain that is structurally and relationally produced (Doyle Fosco, 2022). In contrast, this study adopts a relational ecological perspective, conceptualising well-being as co-constructed across interdependent personal, relational, and organisational levels.
At the personal level, well-being encompasses emotional regulation, psychological vitality, and sustained professional purpose. At the relational level, it is shaped by trust, psychological safety, and shared responsibility. At the organisational level, it is influenced by workload structures, decision-making autonomy, and access to support. Alignment across these levels enables the distribution of emotional labour and sustains leadership capacity over time, while misalignment generates vulnerability, exhaustion, and withdrawal.
This ecological framing resonates with research on resilient school leadership that conceptualises sustainability as relationally and organisationally embedded rather than individually possessed (Day and Gu, 2013). It also aligns with insights from positive organisational scholarship, which emphasise that well-being emerges through relational systems, shared meaning, and organisational routines rather than isolated individual capacities (Truss et al., 2013). Within this perspective, well-being is embedded in everyday organisational practices and relational norms rather than treated as an individual attribute leaders must cultivate in isolation.
The relational job demands–resources model
Integrating these theoretical strands, the study advances a relational JD–R model in which relational trust operates as both a moderator and mediator within the demands–resources–well-being relationship. Job demands exert direct negative pressure on principal well-being, while job resources and relational trust exert direct positive effects. Crucially, relational trust moderates the relationship between job demands and well-being by reshaping how leadership pressure is experienced, buffering emotional strain under conditions of high trust and amplifying vulnerability under conditions of low trust.
In addition, principal well-being is conceptualised as a mediating mechanism linking organisational conditions to broader school-level outcomes. Leaders’ psychological functioning influences their relational availability, emotional regulation, and consistency in leadership practice, thereby shaping school climate and teacher trust. In this framework, relational trust represents the relational quality of interactions among school actors, whereas teacher trust and school climate represent broader organisational outcomes that may be influenced by leadership processes and leader well-being. This positioning aligns with research demonstrating that leadership efficacy and organisational influence are closely connected to leaders’ psychological resources and relational capacity (Gorrell and De Nobile, 2023; Leithwood and Jantzi, 2008).
In this sense, well-being functions not merely as an outcome of organisational conditions but as a relational conduit through which leadership capacity is sustained or eroded over time. Based on this framework, the study advances the following hypotheses:
Job resources and relational trust will positively predict principal well-being.
Job demands will negatively predict principal well-being.
Relational trust will moderate the relationship between job demands and principal well-being.
Principal well-being will mediate the effects of job demands, job resources, and relational trust on perceived school climate and teacher trust.
Methodology
Research design
This study employed a sequential explanatory mixed-methods design (Creswell and Plano Clark, 2018; Fetters et al., 2013) to examine how job demands, job resources, and relational trust interact to shape principal well-being in Cypriot primary schools. This design was selected because it enables the integration of broad quantitative patterns with in-depth qualitative explanations, thereby providing a comprehensive understanding of leadership well-being as both an organisational and relational phenomenon. The quantitative phase tested theoretically derived hypotheses grounded in the JD–R model (Bakker and Demerouti, 2017), while the qualitative phase explored the relational and organisational processes underlying the statistical relationships. This design is well suited to educational management research, where leadership experiences are embedded within complex social systems.
Although principal well-being is conceptualised in this study as relationally and organisationally embedded, the empirical analysis focuses on principals’ individual perceptions of organisational conditions. This approach reflects the logic of the JD–R framework, which examines how structural and relational characteristics of work are experienced at the individual level. Accordingly, constructs such as job demands, job resources, and relational trust are operationalised through principals’ perceptions of their organisational environments. While this design allows examination of how organisational conditions are experienced by school leaders, future research could extend this approach through multi-informant or multilevel designs incorporating teacher or staff perspectives in order to capture more explicitly the collective dimensions of leadership well-being.
Quantitative phase
Participants and sampling
The quantitative phase targeted the population of public primary school principals in Cyprus (approximately 340 principals). A census-based sampling approach was adopted to maximise representativeness and minimise sampling bias. A total of 232 principals completed the questionnaire, yielding a response rate of approximately 66%, which is comparable to response rates reported in survey-based studies involving school leaders, particularly in high-demand leadership contexts (Kaufman et al., 2022).
The sample included principals from urban (60%), semi-urban (25%), and rural (15%) schools and reflected a balanced gender distribution (56% female, 44% male). Participants’ mean leadership experience was 9.8 years (SD = 1.2), closely aligned with national leadership demographics, supporting the relevance of the sample for examining principal well-being in Cyprus. To assess potential non-response bias, sample characteristics were compared with national distributions across gender, school location, and years of leadership experience, revealing close alignment. While response bias cannot be fully excluded, the demographic representativeness of the sample supports the analytical validity of the findings. Nevertheless, results should be interpreted with appropriate caution.
Instruments
Data were collected using a structured questionnaire comprising established and validated instruments. Principal well-being was measured using the Warwick–Edinburgh Mental Well-Being Scale (WEMWBS; Tennant et al., 2007), a 14-item scale assessing positive psychological functioning and affect. JD–R were assessed using adapted items from the Job Demands–Resources Questionnaire for Educators (Bakker and Demerouti, 2017), capturing workload, time pressure, autonomy, and collegial support. Relational trust was measured using the Omnibus T-Scale (Hoy and Tschannen-Moran, 2003), which assesses trust in colleagues, parents, and students. School-level outcomes, including teacher trust and school climate, were measured using validated scales reported in Collie (2022). All items were rated on a five-point Likert scale (1 = Strongly disagree to 5 = Strongly agree). To ensure linguistic and conceptual equivalence, all instruments were translated into Greek and back-translated into English by independent bilingual translators. Pilot testing with a small group of principals confirmed item clarity and contextual relevance.
Data collection and ethics
Quantitative data were collected online during the 2024–2025 academic year. Ethical approval was obtained from the Cyprus Ministry of Education Ethics Committee. Participation was voluntary, informed consent was obtained electronically, and anonymity was assured. Participants were informed of their right to withdraw at any stage without consequence.
Data analysis
Data analysis was conducted using SPSS 29 and AMOS 29. Initial analyses included descriptive statistics and reliability testing using Cronbach's alpha. Confirmatory factor analysis (CFA) was conducted to assess construct validity and discriminant validity. Structural equation modelling (SEM) using maximum-likelihood estimation was then employed to test the hypothesised relationships among job demands, job resources, relational trust, and principal well-being. Model fit was evaluated using established criteria (CFI > 0.90, TLI > 0.90, RMSEA < 0.08).
To examine indirect and interaction effects, mediation and moderation analyses were conducted within the SEM framework. The moderating role of relational trust was tested using interaction terms, while indirect effects were assessed using bootstrapping with 5000 resamples, consistent with best practice in organisational research.
Because the quantitative phase relied on self-reported survey data collected from a single respondent group, the possibility of common method bias was considered. Several procedural steps were implemented to mitigate this concern, including the use of established and validated measurement instruments, assurances of anonymity, and voluntary participation to reduce social desirability effects. In addition, the questionnaire design separated predictor and outcome constructs conceptually and psychologically through the use of distinct scale sections. While these procedures reduce the likelihood that common method variance fully explains the observed relationships, the possibility cannot be entirely excluded and should be considered when interpreting the results. In addition to these procedural remedies, a statistical assessment of common method bias was conducted using Harman's single-factor test. An exploratory factor analysis of all measurement items revealed that the first unrotated factor accounted for 34.2% of the total variance, which is below the commonly accepted threshold of 50% (Podsakoff et al., 2003). This suggests that common method variance is unlikely to pose a serious threat to the validity of the findings.
Qualitative phase
Participant selection
The qualitative phase involved eight principals purposefully selected from the survey respondents to capture variation in reported well-being levels. Selection was based on scores from the WEMWBS, ensuring representation of high, moderate, and low well-being profiles. The sample reflected diversity in gender, school size, and geographical setting, enhancing the analytic depth of the qualitative findings.
Participants were selected based on variation in reported well-being rather than on prior measures of relational trust. During the qualitative analysis, however, clear patterns emerged in the ways principals described relational dynamics within their schools. These patterns enabled the identification of contexts characterised by relatively stronger or weaker trust relationships. References to higher trust or lower trust school contexts in the findings therefore reflect interpretive patterns that emerged inductively from interview narratives, reflective journals, and documentary evidence rather than predefined sampling categories.
Data collection
Qualitative data were collected through semi-structured interviews, 2-week reflective journals, and document analysis (e.g. staff meeting minutes, school improvement plans, newsletters). Interviews lasted between 60 and 90 min and explored principals’ experiences of job demands, relational trust, emotional labour, and organisational support. Reflective journals captured real-time emotional responses and sense-making processes, while document analysis provided contextual validation of reported practices.
Data analysis
Qualitative data were analysed using reflexive thematic analysis (Braun and Clarke, 2021). Analysis proceeded through iterative phases of familiarisation, initial coding, theme development, and refinement. Coding was primarily inductive, allowing patterns to emerge from participants’ accounts, while being theoretically sensitised by concepts from the JD–R framework and relational trust literature. This sensitisation informed interpretation without imposing predefined categories.
Analysis was conducted by the first author, consistent with a reflexive approach that emphasises researcher engagement and interpretive responsibility rather than inter-coder reliability as a marker of quality. Rigour was supported through prolonged engagement with the data, reflexive memo-writing, and triangulation across multiple qualitative sources, including interviews, reflective journals, and documentary materials. This triangulation enabled the comparison of principals’ reported experiences with documented organisational practices and contextual school information.
Integration of quantitative and qualitative phases
Integration occurred primarily at the interpretive level, with qualitative findings used to explain and deepen understanding of the quantitative results (Fetters et al., 2013). Specifically, this integration involved linking the quantitatively identified moderation effect of relational trust with qualitative accounts of shared responsibility, openness, and collective problem-solving, which illustrated how trust reshaped the experience of leadership demands in practice. In addition, the mediating role of principal well-being in connecting organisational conditions to school-level outcomes was interpreted through qualitative evidence highlighting principals’ relational availability, emotional regulation, and influence on school climate and teacher trust. This integration enabled the identification of relational mechanisms, such as trust-based emotional containment and collective sense-making, that clarified how job demands and resources were associated with differential well-being outcomes across school contexts. The combined analysis strengthened the study's explanatory power and informed the development of the proposed Relational JD–R model.
Results
Quantitative findings
Descriptive statistics
The analysis began by examining overall levels of well-being among Cypriot primary school principals. Scores on the WEMWBS indicated moderate-to-high levels of psychological well-being (M = 3.74, SD = 0.58 on a five-point scale). Variation across school contexts was evident. Principals working in smaller rural schools reported significantly higher well-being (M = 3.91, SD = 0.47) than those in larger urban schools (M = 3.59, SD = 0.62), t (230) = 2.25, p < 0.05, suggesting that organisational scale and contextual complexity may shape leadership strain. Gender differences were small and non-significant (female: M = 3.69; male: M = 3.79; p > 0.05).
Overall, principals reported high job demands (M = 3.92, SD = 0.53), reflecting substantial administrative workload, accountability pressure, and emotional labour. Job resources were rated at moderate-to-high levels (M = 3.68, SD = 0.64), indicating perceived autonomy and collegial support alongside systemic constraints. Relational trust emerged as a notable strength, demonstrating high internal consistency (α = 0.88) and a strong positive correlation with job resources (r = 0.71, p < 0.001), suggesting close alignment between supportive organisational conditions and trust-based relationships.
Measurement model and structural model fit
Confirmatory factor analysis (CFA) was conducted to assess the measurement properties of the four latent constructs: job demands, job resources, relational trust, and principal well-being. All items loaded significantly on their respective factors (p < 0.001), with standardised factor loadings ranging from 0.62 to 0.88, exceeding recommended thresholds. Composite reliability values ranged from 0.81 to 0.90, indicating satisfactory internal consistency. Average variance extracted (AVE) values exceeded 0.50 for all constructs, supporting convergent validity. Discriminant validity was confirmed, as the square root of AVE for each construct exceeded its correlations with other constructs. Overall, the measurement model demonstrated acceptable fit and supported the use of the constructs in subsequent analyses.
The structural model demonstrated acceptable fit to the data (χ2/df = 2.31, CFI = 0.94, TLI = 0.93, RMSEA = 0.072, SRMR = 0.061), indicating that the proposed model adequately represented the observed relationships. The model explained 58% of the variance in principal well-being (R2 = 0.58), reflecting substantial explanatory power for organisational research.
Direct effects
SEM revealed relationships consistent with the hypothesised model. Job demands exerted a significant negative effect on principal well-being (β = −0.43, p < 0.001), supporting H2. Principals experiencing higher workload intensity, policy pressure, and emotional demands reported lower levels of psychological vitality.
In contrast, job resources were positively associated with well-being (β = 0.46, p < 0.001), supporting H1. Greater autonomy, collegial support, and access to professional resources were linked to higher well-being. Relational trust also demonstrated a strong positive association with principal well-being (β = 0.39, p < 0.001), further supporting H1 and indicating that trust contributed uniquely beyond structural resources.
Moderating role of relational trust
The interaction between job demands and relational trust was statistically significant and positive (β = 0.24, p < 0.01), supporting H3. This finding indicates that relational trust moderated the relationship between job demands and principal well-being. Simple slope analyses showed that under conditions of high trust (1 SD above the mean), the negative association between demands and well-being was substantially weakened (b = −0.20, p > 0.05). Conversely, under low-trust conditions (1 SD below the mean), job demands were strongly and negatively associated with well-being (b = −0.60, p < 0.001). These results suggest that relational trust operates as an organisational buffer, shaping how leadership demands are experienced rather than reducing their objective intensity.
Mediating role of principal well-being
Mediation analyses indicated that principal well-being functioned as a key mechanism linking organisational conditions to school-level outcomes. Job resources exerted a positive indirect effect on perceived school climate through principal well-being (β_indirect = 0.28, SE = 0.08, 95% CI [0.12, 0.44]), while relational trust exerted an indirect effect on teacher trust via principal well-being (β_indirect = 0.31, SE = 0.09, 95% CI [0.15, 0.47]). Bootstrapped confidence intervals excluded zero for the indirect effects of job resources and relational trust, supporting these mediation pathways. However, the indirect effects of job demand on school-level outcomes via principal well-being were weaker and not consistently statistically significant across all outcomes. Accordingly, the findings provide partial support for H4.
Collectively, the quantitative findings indicate that principal well-being is shaped by the interaction of job demands, organisational resources, and relational trust. While job demands were consistently associated with reduced well-being, both job resources and relational trust emerged as protective conditions, with trust in particular moderating how leadership pressure was experienced. Mediation analyses further suggest that principal well-being operates as a relational mechanism through which organisational conditions influence broader school outcomes, including school climate and teacher trust, although these effects were more consistent for job resources and relational trust than for job demands. However, the quantitative results alone cannot fully explain how similar levels of demand produce divergent well-being outcomes across school contexts. To address this question, the qualitative phase explores the organisational and relational processes through which trust, shared responsibility, and meaning-making shape principals’ everyday experiences of leadership strain and sustainability.
Qualitative findings
The qualitative phase was designed to explain and elaborate the quantitative relationships identified in the structural model, particularly the buffering role of relational trust and the mediating role of principal well-being. Analysis of interviews, reflective journals, and school documents yielded three interrelated themes that clarify how similar structural conditions produced markedly different well-being outcomes across school contexts.
The relentless intensity of leadership work
Across all cases, principals described leadership as continuous, intense, and emotionally demanding, confirming the high levels of job demands identified quantitatively. However, intensity alone did not determine well-being outcomes. Instead, principals’ interpretations of pressure, shaped by relational context, proved decisive. Principals reporting higher well-being framed workload pressure as meaningful and collectively manageable. Leadership demands were described as inherent to the role but softened through shared responsibility and moral purpose. As one principal explained:
The work is heavy, but it doesn’t sit on me alone. When teachers take responsibility and we solve problems together, the pressure becomes something we manage as a team rather than something that belongs only to the principal. (Principal 3)
There are always ministry directives, parents’ issues, and staff concerns happening at the same time. But when the staff trust each other, the problems move through the school rather than stopping at my office. (Principal 7)
Everything comes back to me in the end. You feel that you must always appear capable and calm, even when you are overwhelmed. There is no real space to say that something is difficult. (Principal 2)
Relational trust as emotional infrastructure
Relational trust emerged as the central mechanism shaping how principals processed and managed leadership pressure. In high-trust schools, trust functioned as emotional infrastructure, enabling openness, psychological safety, and informal support. Principals described everyday relational practices, including brief check-ins, shared humour, and collective reflection, as mechanisms that absorbed pressure and prevented emotional overload. As one principal reflected:
When trust is there, problems don’t disappear, but they stop feeling personal. You know that teachers will support you and that you can speak honestly about challenges. (Principal 5)
If a difficult situation appears, we talk about it openly. Teachers bring ideas and sometimes they solve things faster than I could alone. That makes the work feel shared. (Principal 6)
You have to appear in control all the time. If you show uncertainty, people may question your leadership, so you carry many things silently. (Principal 1)
Well-being as a shared organisational responsibility
The final theme clarifies the mediating role of principal well-being between organisational conditions and school-level outcomes. Principals with higher well-being consistently framed well-being as a collective organisational responsibility embedded in leadership routines and school culture. Practices such as recognising staff effort, distributing leadership, and creating space for emotional expression were described as intentional strategies sustaining both leader and staff energy. As one principal observed:
If I am depleted, the school feels it immediately. Teachers notice the tone, the energy, everything. That is why I try to protect the emotional climate of the school. (Principal 4)
When people feel supported, they support each other. That creates a different atmosphere in the school, and the pressure becomes easier to manage. (Principal 8)
You learn to accept that being tired is part of the job. There is always something urgent, and after some time you stop expecting it to change. (Principal 2)
Discussion
This study set out to examine principal well-being as an organisational and relational phenomenon by integrating the JD–R framework (Bakker and Demerouti, 2017; Demerouti et al., 2001) with theories of relational trust in schools (Bryk and Schneider, 2002; Hoy and Tschannen-Moran, 2003). Across quantitative and qualitative phases, the findings converge on a central insight relevant to educational management scholarship: while job demands remain a persistent feature of contemporary school leadership, their impact on principal well-being appears to vary substantially depending on relational conditions, particularly trust. In this sense, well-being emerges not merely as an individual psychological state but as an organisational condition embedded in everyday leadership practice.
Reframing leadership strain: job demands as structurally persistent but relationally Variable
Consistent with JD–R theory, job demands were found to be negatively associated with principal well-being, reflecting the sustained intensity of contemporary leadership work (Bakker and Demerouti, 2017; Kaufman et al., 2022). However, the mixed-methods findings indicate that although demands were broadly consistent across contexts, well-being outcomes varied considerably among principals. This pattern suggests that leadership strain may not be fully explained by workload intensity alone, supporting critiques of demand-centred explanations of principal burnout that overlook organisational and relational mediation. This variation is further clarified by the quantitative finding that relational trust moderates the relationship between job demands and well-being, indicating that the impact of demands is contingent on relational conditions rather than fixed in intensity.
The qualitative findings illuminate this divergence by showing that the meaning principals attributed to leadership pressure was shaped by relational context. Principals working in higher trust environments described demands as collectively shared and morally purposeful, whereas those in lower trust settings experienced similar pressures as isolating, evaluative, and emotionally depleting. These accounts provide an explanatory lens for the moderation effect, illustrating how trust reshapes the subjective experience of leadership demands through shared responsibility, openness, and collective sense-making. These findings align with educational leadership research suggesting that leadership isolation, rather than demand volume alone, constitutes a significant risk factor for diminished leader well-being (Dor-Haim and Oplatka, 2021). From this perspective, job demands may operate less as fixed stressors and more as relationally mediated experiences whose psychological impact depends on organisational conditions.
Relational trust as a meta-resource buffering leadership strain
A central contribution of this study lies in providing empirical evidence supporting the moderating role of relational trust within the JD–R framework. While trust has long been associated with collaboration, organisational learning, and school improvement (Bryk and Schneider, 2002; Collie, 2022), its role in shaping principals’ own experience of leadership strain has received comparatively limited empirical attention. The present findings suggest that relational trust does not simply coexist with well-being but may shape how leadership demands are interpreted and managed by principals.
Quantitative moderation analyses indicated that higher levels of relational trust weakened the negative association between job demands and principal well-being. Qualitative evidence helps clarify the mechanisms underlying this relationship, showing that trust fostered psychological safety, relational openness, and shared responsibility. In higher trust schools, principals described leadership challenges as collective organisational concerns rather than individual burdens, potentially reducing emotional overload. Conversely, lower trust environments constrained vulnerability and help-seeking, intensifying the personal cost of leadership work (Bukko et al., 2021).
These dynamics are consistent with conservation of resources theory (Hobfoll, 2011), which conceptualises resources as operating through cycles of gain and loss. Within this perspective, trust may function as a meta-resource that amplifies the protective capacity of other resources and mitigates resource loss. Evidence from Cypriot primary schools further indicates that trust-based relational practices may partially compensate for structural constraints by enabling emotional containment and distributed responsibility (Keravnos et al., 2025). This finding extends existing trust scholarship by suggesting that relational trust may operate not only as an organisational asset but also as a stabilising condition for leadership sustainability.
Rethinking job resources through relational accessibility
The strong positive association between job resources and principal well-being supports established JD–R findings in educational leadership (Chen et al., 2025). However, this study refines existing interpretations by suggesting that the effectiveness of job resources may depend on relational accessibility. Autonomy, collegial support, and professional learning opportunities appeared to function as meaningful resources primarily when embedded within trust-based organisational conditions.
Qualitative accounts revealed that in lower trust contexts, formally available resources were sometimes experienced as conditional or risky. Support could be interpreted as evaluative rather than relational, delegation as exposure rather than empowerment, and autonomy as constrained by surveillance. In contrast, higher trust environments appeared to transform similar structural resources into sources of motivation and psychological relief (Tschannen-Moran, 2014). These findings reinforce arguments within educational management scholarship advocating for moving beyond purely structural conceptions of leadership support toward relationally embedded understandings of how demands and resources are experienced in practice (Tummers and Bakker, 2021). For policy and practice, this suggests that resource provision alone may be insufficient without parallel investment in relational infrastructure.
Principal well-being as a relational mechanism shaping organisational outcomes
The mediation analyses further highlight the organisational significance of principal well-being. Rather than functioning solely as an individual outcome, well-being appears to operate as a relational mechanism linking organisational conditions to school-level outcomes, including school climate and teacher trust. Principals reporting higher well-being described greater relational availability, emotional regulation, and consistency in leadership interactions; qualities that may plausibly influence organisational tone and staff perceptions.
Conceptualising principal well-being as a mediating mechanism aligns with leadership research suggesting that leaders influence organisational outcomes primarily through indirect and relational processes rather than direct control (Hallinger and Heck, 2010). The findings also resonate with research documenting cascading effects of leader well-being on organisational functioning, including morale, trust, and stability (Collie, 2022; Gorrell and De Nobile, 2023). By conceptualising well-being as an intermediary process rather than an endpoint, the study highlights its potential strategic relevance for educational management.
From individual resilience to a relational ecology of well-being
Taken together, the findings support a shift away from purely individualised resilience approaches toward what may be described as a relational ecology of well-being. Within this perspective, leadership sustainability appears to be co-produced through structural arrangements, trust-based relationships, and shared meaning-making practices. While individual coping strategies remain relevant, the study suggests that well-being was most sustainable where emotional labour was implicitly distributed, vulnerability was normalised, and leadership challenges were framed collectively.
This ecological perspective resonates with research on resilient school leadership that conceptualises sustainability as relationally and organisationally embedded rather than individually possessed (Day and Gu, 2013). It also responds to critiques of well-being interventions that place disproportionate responsibility on individual leaders while leaving organisational conditions unchanged (Doyle Fosco, 2022). Within educational management scholarship, this reframing positions well-being as a core organisational concern rather than a peripheral or personal one.
Transferability and international relevance
Although this study is situated within the Cypriot primary education system, its contribution lies less in statistical generalisation and more in theoretical transferability. Cyprus represents a highly centralised, high-accountability context characterised by limited principal autonomy, dense administrative regulation, and close relational proximity among school actors (Pashiardis and Johansson, 2021; Savvides and Pashiardis, 2016). These features intensify leadership demands while amplifying the salience of relational conditions, rendering the context theoretically informative for examining how leadership well-being is sustained under pressure.
The relational mechanisms identified, particularly the buffering role of trust in shaping the experience of leadership demands, may also resonate with findings from other high-demand systems. Research in more decentralised contexts, including the United States and the United Kingdom, similarly documents how trust, relational support, and collective sense-making may mitigate leadership strain during periods of heightened accountability and crisis (Kaufman et al., 2022).
Rather than suggesting direct applicability across settings, this study offers a theoretically grounded illustration of how relational infrastructure interacts with JD–R to shape leadership well-being. The proposed relational JD–R model therefore provides an analytical lens that may be explored and adapted in other systems facing intensifying accountability pressures and leadership sustainability challenges.
Limitations
While this study offers a theoretically informed and empirically rich account of principal well-being in Cypriot primary schools, several limitations should be acknowledged.
First, the cross-sectional design of the quantitative phase limits causal inference. Although the relationships among job demands, job resources, relational trust, and well-being were theoretically grounded and statistically robust, the data do not permit firm conclusions regarding directionality. Leadership well-being is likely to be dynamic, shaped by policy change, organisational conditions, and relational processes over time. Longitudinal designs would enable future research to examine temporal patterns, resource gain and loss cycles, and the durability of trust-based buffering effects more rigorously.
Second, the study relied primarily on self-reported measures, which may introduce social desirability bias or subjective inflation. Although the mixed-methods design and qualitative triangulation strengthened interpretive validity, and both procedural and statistical steps were taken to mitigate common method bias, future research could incorporate additional data sources, such as peer assessments, supervisor evaluations, or organisational indicators, to provide a more comprehensive assessment of leadership well-being and its organisational correlates.
Third, the qualitative phase involved a purposeful sample of eight principals, limiting statistical generalisability. However, the aim of the qualitative inquiry was analytic depth rather than representativeness, enabling close examination of relational processes underlying the quantitative patterns. Further research across a wider range of schools and leadership contexts would be valuable for examining the transferability of the proposed relational ecology of well-being.
Finally, the contextual specificity of the Cypriot education system should be considered when interpreting the findings. Cyprus's small scale and highly centralised governance structure shape the salience of trust, relational density, and leadership demands, conditions that may differ from those in larger or more decentralised systems. Accordingly, the findings should be understood as contextually grounded rather than universally prescriptive, contributing to theory-building rather than direct generalisation.
Suggestions for further research
Building on the present findings, several directions for future research emerge.
Longitudinal studies are needed to examine how principal well-being evolves across school years, reform cycles, or periods of crisis. Such designs would enable investigation of resilience trajectories, recovery processes, and the durability of relational trust as a buffering mechanism over time, providing a more dynamic test of the relational JD–R model.
Intervention-focused research could examine the causal impact of organisational and relational initiatives, such as trust-building practices, distributed leadership structures, or workload redesign, on principal well-being. Quasi-experimental or design-based approaches would be particularly valuable in identifying which organisational levers most effectively sustain leadership capacity under high demand.
Comparative cross-system studies could explore whether the relational ecology of well-being identified in Cyprus operates similarly across contexts with different governance structures, cultural norms, or accountability regimes, clarifying the extent to which trust functions as a contextually contingent resource.
Future mixed-methods research could adopt a multi-stakeholder perspective, incorporating teacher and middle-leader voices to examine how principal well-being interacts with staff morale, trust, and collective efficacy, further reinforcing its conceptualisation as a systemic rather than individual condition.
Finally, emerging developments such as digitalisation and the use of artificial intelligence in school administration warrant closer examination. Future studies could investigate how technological mediation reshapes leadership demands, emotional labour, and trust dynamics, introducing new risks and resources within the relational ecology of well-being.
Implications
The findings of this study carry important implications for educational management practice and policy, particularly in systems characterised by high accountability and constrained autonomy. First, the results highlight the potential value of reframing principal well-being as a strategic organisational condition rather than solely a private or personal concern. From an educational management perspective, leader well-being appears closely connected to organisational stability, relational climate, and leadership sustainability. Accordingly, system-level policies and leadership frameworks may benefit from recognising principal well-being as an integral component of effective school management.
Second, the findings suggest that organisational responses to leadership strain may need to extend beyond individualised stress management or resilience-based interventions. While such approaches may offer short-term relief, they risk individualising responsibility for pressures that are structurally and relationally produced. The evidence presented in this study indicates that organisationally embedded supports, such as structured peer networks, mentoring arrangements, and facilitated spaces for collective reflection, may help reduce leadership isolation and distribute emotional labour, particularly when these practices are grounded in trust-based rather than evaluative norms.
Third, relational trust may be understood as a core managerial resource rather than merely an interpersonal attribute. The findings indicate that trust shapes how demands and resources are experienced and mobilised in practice. Leadership preparation and professional development programmes may therefore benefit from placing greater emphasis on relational work, emotional labour, and trust cultivation as components of effective educational leadership. Developing leaders’ capacity to foster openness, psychological safety, and shared responsibility could contribute to more sustainable leadership environments.
Finally, the findings highlight the limits of relational buffering in the absence of structural alignment. Although trust can mitigate the subjective impact of leadership demands, it cannot fully compensate for persistent structural pressures such as excessive administrative burden, unclear policy communication, or constrained decision-making authority. Policymakers may therefore need to consider how organisational structures influence leadership sustainability. Addressing workload intensification and creating space for distributed leadership and professional judgement could represent important conditions for supporting principal well-being in demanding educational systems.
Conclusion
This study contributes to educational leadership and management scholarship by conceptualising principal well-being as a relational and organisational phenomenon rather than solely an individual psychological trait. Integrating quantitative and qualitative evidence, the findings indicate that while job demands remain a persistent feature of contemporary school leadership, their impact on well-being appears to be shaped, and in some cases moderated, by organisational resources and relational trust. By extending the JD–R framework to incorporate relational trust as both a moderating and mediating mechanism, the study refines existing theoretical perspectives and advances a relational ecology of well-being as an analytical lens for understanding leadership sustainability. This perspective highlights how well-being may be co-produced through everyday relational practices, shared meaning-making, and trust-based organisational conditions embedded within school life.
Taken together, the findings suggest that sustaining school leadership under conditions of high demand cannot be fully understood through individual endurance or resilience alone. Instead, leadership well-being appears to emerge as a relational and organisational condition shaped by the interaction of structural demands, available resources, and trust-based relationships. By empirically integrating relational trust into the JD–R framework, this study offers a contextually grounded contribution to theory-building on leadership well-being, illustrating how relational infrastructure may reshape the experience of leadership pressure in a highly centralised system. While the findings are not intended to be universally generalisable, they provide theoretically informative insights into the relational mechanisms through which leadership sustainability may be supported in demanding organisational contexts.
As education systems continue to confront intensifying accountability pressures and emotional demands, understanding well-being as relationally embedded rather than solely individually managed becomes increasingly important for sustainable educational leadership.
Footnotes
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.