A Cluster-Analytic Approach to Preschool Teachers’ Psychological and Behavioral Profiles: Irrational Beliefs, Burnout, and Innovative Work Behavior

Abstract

Individual beliefs are a critical factor in understanding human action and behavior. Certain beliefs, such as irrational beliefs and burnout, influence all forms of learning and social interaction within the school environment, primarily limiting both individual and collective development. The former are associated with the inherent human tendency to adhere to habits and behaviors not strictly dictated by rationality, often stemming from irrational thoughts held by the individual. The latter, examined within the framework of the Job Demands–Resources Theory, pertain to occupational characteristics that differentially affect employees’ well-being, job demands, and available resources. The present study aims to investigate the role of these variables in relation to teachers’ Innovative Work Behavior, a recurring, multi-stage process oriented toward the implementation of new ideas within the school context. The sample consisted of 337 preschool educators who completed self-report questionnaires. Multiple linear regression analysis indicated that both irrational beliefs (positively) and the dimension of work disengagement (negatively) significantly influenced innovative work behavior, underscoring the distinct contributions of personal belief systems and burnout dimensions. Furthermore, a hierarchical cluster analysis revealed both heterogeneity among educators and common, distinct response patterns. The identification of five different clusters suggests that the examined characteristics and the underlying beliefs represent individual traits that change dynamically, leaving open the possibility of nonlinear relationships present in the workplace. Five profiles were identified, namely Disengaged-Low Innovators, Resilient-Balanced Innovators, Adaptive Innovators, Strained but Innovative Innovators, and Belief-Driven Innovators, which highlight the complex ways in which disengagement, exhaustion, and irrational beliefs combine to shape innovative work behavior. The findings are interpretable within the framework of contemporary theories in organizational psychology and management and can be utilized by educational principals to enhance school climate and teacher performance.

1. Introduction

Education constitutes the cornerstone of social development and prosperity, since the cultivation of critically minded individuals, together with the provision of knowledge and the dissemination of ideas, appears to be pivotal in addressing the increasing and complex demands of contemporary societies. The rapid advancement of technology and knowledge, the multicultural character of schools, the interventions and expectations of parents, as well as the role of leadership, all highlight the pressing need for the integration of innovation into educational practice. Central to this endeavor are teachers, whose attitudes and beliefs form the primary focus of the present dissertation.

Driven by the pursuit of transcending conventional practices, professionals in education are called to engage in a cyclical and multi-stage process aimed at the adoption and promotion of novel ideas oriented towards collective benefit. A prerequisite for this process is the cultivation of Innovative Work Behavior (IWB), which has been conceptualized as a dynamic construct comprising idea generation, idea promotion, and idea realization (Lambriex-Schmitz et al., 2020; Messmann & Mulder, 2020). While it is widely acknowledged that IWB in the workplace contributes to organizational advancement, it is not yet a broadly diffused characteristic among educators. The production and dissemination of alternative ideas do not necessarily guarantee their universal acceptance or successful implementation, as individual and contextual factors can either facilitate or hinder innovation. Consequently, teachers’ personal resources, as well as the demands they face, play a decisive role in shaping the flourishing, or stagnation, of IWB (Kurniawan et al., 2025; Liu & Sun, 2025; Mokhlis & Abdullah, 2024).

An important dimension in this discussion concerns the presence of Irrational Beliefs (IB), which are often internalized beliefs that inhibit adaptive functioning and exert non-constructive consequences on individuals’ actions and lifestyles. IB represent structured perceptions of reality, shaped by everyday experiences, and they account for different manifestations of behavior (David et al., 2018). Such beliefs often operate as predispositions that bypass rational processing, reinforcing either positive or negative biases in teachers’ professional lives. As Ellis (1997) argued, these irrational cognitive patterns are frequently detached from logical reasoning and can influence decision-making in ways that limit openness to innovation or resilience in the face of challenges. Recent evidence confirms that IB, such as rigid demands and low frustration tolerance, are linked to academic stress and maladaptive coping (Turner et al., 2024) and that they can significantly predict burnout among teachers (Benigno et al., 2025; Huk et al., 2019).

The school environment, moreover, is a fertile ground for work-related stress, which in turn may culminate in Burnout, which has been conceptualized as the outcome of increased job demands coupled with insufficient job resources, reflecting a chronic imbalance between what is required of individuals and the means available to them (Bakker & Demerouti, 2014). This dual perspective underscores the interplay between environmental pressures and personal coping capacities, highlighting how sustained strain without adequate support mechanisms may erode both professional efficacy and personal well-being. It often undermines educators’ ability to adapt to professional demands and leads to decreased performance, indifference, and diminished motivation. Importantly, teacher burnout does not remain confined to the individual but reverberates throughout the entire educational organization. It negatively impacts all members of the school community, shaping both learning and social activities, and can ultimately provoke collective dysfunction (Shukla & Trivedi, 2008; Skaalvik & Skaalvik, 2023).

Thus, the interplay between innovation, cognitive predispositions, and occupational stress underscores the complexity of fostering IWB in education. Teachers are not only knowledge transmitters but also key agents of change; their capacity to embrace innovation depends on both intrapersonal factors, such as overcoming IB, and systemic conditions, such as supportive leadership and manageable workload (Nwoko et al., 2023). Addressing these dimensions holistically is essential if schools are to evolve into environments that not only respond to but also anticipate societal challenges. Although teachers’ IWB has been widely studied, the combined effects of IB and burnout and the identification of distinct psychological profiles among teachers remain underexplored, a gap addressed in the present study.

1.1. Innovative Work Behavior

The distinction between creativity and innovation is central in literature, with creativity defined as the generation of novel ideas and innovation as their systematic refinement and implementation (Scott & Bruce, 1994). Innovation is thus considered a dynamic, iterative, and nonlinear process that requires institutional embedding to ensure long-term organizational sustainability (Gannaway et al., 2013). This perspective has informed the study of IWB, conceptualized as the behavioral framework through which employees contribute to innovation by identifying opportunities, generating ideas, promoting them, and ensuring their implementation (De Jong & Den Hartog, 2010). Later models added reflection (Messmann & Mulder, 2012) and sustainability (Lambriex-Schmitz et al., 2020) as integral dimensions, highlighting that innovation is not complete until new practices are both optimized and stabilized within organizational structures. Current perspectives position IWB as a repetitive, multi-stage process that encompasses overlapping phases. Opportunity exploration and idea generation, once treated as separate, are now viewed as closely interconnected. The process as a whole involves generating ideas, promoting and implementing them, reflecting on their effectiveness, and ensuring their sustainable integration within organizational structures. In educational settings, idea generation can manifest as teachers proposing novel ideas, idea promotion as advocating for these ideas among colleagues or stakeholders, idea realization as the practical implementation of new ideas in classrooms, and idea sustainability as embedding successful practices into long-term school policies and routines. In this way, IWB is conceptualized as a behavioral framework through which employees actively contribute to organizational innovation, ultimately enhancing long-term adaptability and growth (Gkontelos et al., 2022).

Research further demonstrates that IWB is shaped by the interdependence of cognitive, behavioral, and contextual processes. Learning emerges as a core driver, enhancing knowledge, promoting self-efficacy, and enabling employees to secure peer support, which is essential for implementing new ideas (Holman et al., 2012; Lecat et al., 2018). At the same time, rapid technological developments and increasing competition push organizations to adapt swiftly, requiring employees to transcend routine tasks and reinterpret challenges as creative opportunities (Carvalho et al., 2023; Zhang & Bartol, 2010). Importantly, managerial relationships, organizational climate, and the absence of fear of failure also play a decisive role, fostering psychological safety and reinforcing IWB (Zhang et al., 2021).

In the educational sector, schools are identified as dynamic environments where IWB is particularly relevant, given the pressures of interdisciplinarity, multiculturalism, social expectations, and the integration of new technologies (Thurlings et al., 2015; Xafakos et al., 2020). Teachers’ IWB is essential not only for their personal and professional development but also for enhancing school-wide innovation capacity (Messmann et al., 2010). This behavior enriches teaching practices, improves collegial collaboration, and contributes to organizational well-being by fostering satisfaction and continuous skill enhancement (Aryani et al., 2024; Messmann et al., 2018; Tuominen & Toivonen, 2011). Thus, IWB serves both as an individual trait of teachers and as an element of school culture, making it a critical determinant of educational effectiveness. It translates into enhanced student outcomes, greater classroom engagement, and improved school-wide effectiveness.

Finally, empirical findings show that, while demographic factors such as gender or education level have limited influence (Bahru et al., 2023), personal and organizational factors are decisive. At the individual level, traits such as curiosity, openness, intrinsic motivation, and creative self-efficacy consistently support IWB (Ghosh et al., 2019; Gkontelos et al., 2023b; Mastrothanasis & Kladaki, 2025). At the organizational level, trust, collaboration, leadership, professional autonomy, and positive school climate are identified as strong enablers (Lin & Liu, 2025; Messmann et al., 2022; Wang et al., 2022). Conversely, rigid curricula, excessive testing, and feelings of isolation are major barriers (Nakata, 2011; Papachristopoulos et al., 2023). Together, these findings confirm that IWB is best understood as a multidimensional, context-sensitive process, shaped by the interplay of individual dispositions and organizational conditions.

1.2. Irrational Beliefs

Human behavior and expression are strongly shaped by the interactions among individual characteristics, with normative, behavioral, and self-regulatory beliefs playing a central role (Ajzen, 1991). Among these, irrational beliefs (IB) are particularly noteworthy, as they manifest across social groups, producing nonlinear patterns and abrupt behavioral changes (Stamovlasis & Vaiopoulou, 2017). Humans frequently deviate from logical behavior, reflecting inherent predispositions independent of cognitive ability or intelligence (Ellis, 1997). IB encompass a wide array of constructs, including dysfunctional myths, schemas, cognitions, and automatic thoughts, highlighting their multifaceted presence across various domains such as education, psychology, psychiatry, and medicine (Bernard, 2016; Ozer & Akgun, 2015; Turner & Bennett, 2018).

Conceptually, IB are evaluative knowledge organized into schemas and propositional representations, capturing structured conceptions of reality and guiding behavior (Anderson, 2000; DiGiuseppe, 1996; Ellis, 2000). Each schema reflects the structure of an object and may assume different values depending on context, while daily experiences can generate or reinforce IB, influencing both behavior and lifestyle. Depending on outcomes, IB are often classified as functional or dysfunctional, emphasizing the consequences of these beliefs on individual actions.

The origins of IB have been debated within the nature versus nurture framework, with prevailing explanations emphasizing the role of education and social-cultural environment (David & DiGiuseppe, 2010), alongside bounded rationality as a conceptual lens (Simon, 1991). IB are unrealistic beliefs lacking empirical support, whose generalization or personalization can result in maladaptive outcomes (David & Cramer, 2010; Ellis et al., 2010). Low IB levels do not automatically imply high functional beliefs (FB), as IB and FB exist as distinct dimensions (David, 2003). Psychometrically, IB combine thought content and psychological processes, influencing cognition, emotion, and behavior in complex ways (Ellis, 2005). Their effects can be categorized as emotional and behavioral, with emotional responses often preceding behavioral outcomes (David & Lynn, 2010).

In educational contexts, IB are particularly relevant for teachers, where ongoing professional challenges and dynamic school environments can exacerbate their development and trigger stress or defensive reactions. Teachers’ IB are commonly grouped into self-downing, authoritarianism, demands for justice, and low frustration tolerance (Bernard, 2016; Gkontelos et al., 2021). Self-downing arises from high personal standards and excessive need for approval and may lead teachers to excessively criticize their own lesson plans or hesitate to implement new teaching strategies due to fear of failure. Authoritarianism reflects responses to student misbehavior, often manifesting as rigid enforcement of rules rather than flexible classroom management approaches. Demands for justice concern expectations for participation in decision-making, causing frustration when teachers perceive unequal involvement or feel their professional input is undervalued. Low frustration tolerance emerges from high workload and expectations for success, which can result in impatience with students’ learning pace or difficulty coping with administrative pressures. These dimensions collectively shape teachers’ perceptions, attitudes, and professional behavior, influencing both classroom interactions and broader school dynamics.

IB influence both emotional and professional outcomes. They are linked to impaired work performance, dysfunctional responses, and lower self-efficacy (Bermejo-Toro & Prieto-Ursúa, 2006; Warren & Hale, 2016; Wood et al., 2017), while also affecting self-esteem, psychosomatic well-being, depression, and stress (Ifelunni et al., 2023; Küçük et al., 2016; Stephenson et al., 2018; Vita-Agundu et al., 2022). IB further correlate with emotional intelligence, perfectionism, professional burnout, job satisfaction, and deviant behavior, yet they negatively impact motivation, teaching performance, and instructional effectiveness (Diale & Victor-Aigbodion, 2022; Ekwueme et al., 2023; Huk et al., 2019; Khaledian et al., 2016; Nwabuko et al., 2020). Consequently, professional development interventions targeting IB are critical, fostering healthy emotional expression, functional behavior, and overall teacher efficacy within the school environment (Ogakwu et al., 2024; Warren & Gerler, 2013).

1.3. Burnout

Teachers’ work environments, characterized by high demands and emotionally charged interactions, often contribute to Burnout, a condition marked by mental exhaustion, disengagement, and cynicism (Goddard et al., 2006; Maslach, 1982). Burnout negatively affects performance, mood, and interpersonal relationships, and is further exacerbated by work-related stress, acting as a barrier to professional development. In educational settings, it impacts not only individual teachers but also the collective functioning of the school, influencing learning processes, social interactions, and organizational cohesion (Shukla & Trivedi, 2008).

Historically, Burnout has been conceptualized as a triad of emotional exhaustion, depersonalization, and reduced personal accomplishment (Maslach et al., 2001). Emotional exhaustion involves fatigue and loss of energy, depersonalization reflects emotional detachment and social isolation, and reduced personal accomplishment relates to diminished beliefs in one’s effectiveness, impacting work performance (Schaufeli et al., 2009). The Job Demands–Resources (JD-R) theory, which is utilized in the present study, offers an alternative framework, defining Burnout as a response to high job demands and insufficient job resources, structured around two dimensions: exhaustion and disengagement (Bakker & Demerouti, 2014; Demerouti et al., 2001).

Exhaustion arises from prolonged emotional, cognitive, and physical strain, while disengagement reflects cynicism, lack of interest, and negative attitudes toward work outcomes (Demerouti et al., 2010). Reduced personal accomplishment is considered a potential consequence of Burnout, akin to self-efficacy (Demerouti et al., 2001). The JD-R model further distinguishes between job demands (e.g., workload, task complexity) and job resources (e.g., social support, feedback), which differently influence well-being. Increased demands contribute primarily to exhaustion, whereas scarce resources lead to disengagement, resulting in fluctuations in performance and motivation (Bakker et al., 2014; Xanthopoulou et al., 2012).

Teachers face multiple job demands including workload, time pressure, role conflicts, interactions with students and parents, and societal expectations (Okeke & Dlamini, 2013). Job resources such as social support, leadership, self-efficacy, autonomy, and emotional intelligence can buffer these demands and prevent Burnout (Geraci et al., 2023; Iacolino et al., 2023; Lucas-Mangas et al., 2022). Empirical studies highlight specific demands affecting exhaustion, including teaching complexity, student diversity, stress, depression, parental involvement, and conflictual climate (Bakker & Demerouti, 2018; Levante et al., 2023). Conversely, resources associated with disengagement encompass feedback, infrastructure, income, motivation, job satisfaction, professional health, and supportive school climate (Martí-González et al., 2023; Shen et al., 2015; Tsang et al., 2022).

Burnout has primarily negative consequences, reducing professional health and the quality of education, and often prompting withdrawal or attrition (Skaalvik & Skaalvik, 2017). Nevertheless, it can also signal the final stage of ineffective stress management, highlighting the potential for targeted interventions and professional development to restore well-being and enhance teachers’ adaptive capacity (Gkontelos et al., 2025; Jennett et al., 2003). Effective management of demands and optimization of resources is thus essential to maintain teacher well-being and sustain school performance. Consequently, teachers experiencing high levels of burnout may provide lower instructional quality, reduced feedback, and diminished classroom engagement, which can negatively affect student learning outcomes. Moreover, burnout dimensions may interact with teachers’ IB and cognitive–emotional resources, influencing the degree to which these factors collectively impact IWB.

2. Materials and Methods

2.1. Current Study and Research Hypotheses

The present study is grounded in the recognition IWB as a variable of considerable research interest and critical relevance to educational practice. The ongoing adaptation of the theoretical framework underscores the necessity for further investigation and theoretical development, which, however, requires the application of robust statistical analyses to ensure the generalizability of findings. The selection of IWB, Burnout, and ΙΒ as focal variables is justified by the observation that teachers’ high cognitive abilities and rational development do not always translate into fully rational judgments, resulting in the formation of erroneous biases. The dual nature of ΙΒ and the variability of their potential responses further necessitate careful examination. Given these considerations and recognizing the need to support the research hypotheses within the chosen theoretical framework, Burnout, within JD-R theory, emerges as an appropriate variable, validating its inclusion in the present study.

Previous research has predominantly employed variable-centered approaches to examine the predictors and outcomes of teachers’ IB and IWB. While these approaches have facilitated understanding of constructing relationships and the development of general intervention strategies, they are limited in assuming population homogeneity and often fail to account for naturally occurring subgroups within the population (Hofmans et al., 2020). This study adopts a person-centered approach, allowing the integration of multiple characteristics simultaneously and providing a broader understanding of their outcomes. Unlike variable-centered methods, person-centered approaches can identify subgroups of teachers with shared attributes, such as low ΙΒ and high IWB, who may benefit from targeted interventions. Specifically, techniques such as hierarchical cluster analysis enable the identification of distinct teacher profiles exhibiting similar patterns of burnout and IB, without relying on predetermined classifications (Morin et al., 2018).

To conclude, the objectives of this study are the following: first, to examine the relationships between teachers’ ΙΒ, IWB, and burnout expecting that teachers who experience higher ΙΒ or burnout will report lower IWB. Moreover, hierarchical cluster analysis will be employed to identify distinct participant profiles, classifying individuals according to their burnout and IB scores. Subsequently, ANOVA will be used to assess whether these clusters differ significantly in terms of their impact on IWB. Finally, an effort will be undertaken to map the extracted profiles in relation to teachers’ IWB, with the aim of identifying potential patterns and underlying associations.

2.2. Participants and Procedures

Data were collected through a self-completion questionnaire administered via LimeSurvey, ensuring participants’ anonymity. An opportunity sampling procedure was employed, relying on teachers who voluntarily chose to participate. School principals across primary schools in Greece facilitated the process by distributing an instructive email to in-service teachers, which included a cover letter outlining this study’s aims, confidentiality assurances, and the voluntary nature of participation. Ethical approval for this study was granted by the institutional Research Ethics and Deontology Committee. The study sample comprised 337 early childhood education teachers, the majority of whom were female (96.4%), with ages ranging from 23 to 54 years (M = 42.97, SD = 8.6). Most participants were employed in kindergartens located in urban areas (city = 50.7%, town = 34.9%, rural = 14.4%), while nearly half (47.5%) held a master’s degree. Teaching experience ranged from 2 to 32 years (M = 14.68, SD = 8.00).

Data were analyzed using JASP software (version 0.19.1.0). Model fit was evaluated using multiple goodness-of-fit indices, following the cutoff criteria proposed by Schermelleh-Engel et al. (2003): Comparative Fit Index (CFI) >0.90 for acceptable fit and > 0.95 for good fit, Tucker–Lewis Index (TLI) > 0.90, Root Mean Square Error of Approximation (RMSEA) < 0.08 for acceptable fit and <0.05 for excellent fit, and Standardized Root Mean Square Residual (SRMR) < 0.08. Factor loadings ≥ 0.40 were taken as evidence of acceptable construct validity, while Cronbach’s alpha and McDonald’s omega values ≥ 0.70 were used to establish internal consistency.

Statistical power and sample size calculations were performed using G*Power software (version 3.1.9.7; Faul et al., 2009). For the multiple linear regression analysis that was used, the required sample size was determined based on an anticipated effect size f2 of 0.15, an alpha error probability of 0.05, and a power of 0.95, indicating that a minimum of 119 participants would be necessary. For the multiple linear regression analysis, a priori power analysis was conducted using G*Power 3.1, assuming a medium effect size (f2 = 0.15), an alpha error probability of 0.05, a desired power of 0.95, and three predictors. The analysis indicated that a minimum of 119 participants was required. Our actual sample (N = 337) therefore exceeded this requirement, ensuring adequate statistical power to detect at least medium-sized effects. For the one-way ANOVA conducted after the hierarchical cluster analysis, an a priori power analysis (α = 0.05, power = 0.95, medium effect size f = 0.25, five groups) indicated that approximately 305 participants would be required. The available sample of 337 participants was therefore sufficient.

2.3. Instruments

Three standardized scales were employed in this study, each assessing one of the three latent variables under investigation. All instruments had been previously adapted for use with the Greek population, either by prior researchers or by the authors. Responses were recorded on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).

The Greek IWB Scale (G-IWBS; Gkontelos et al., 2022) comprises 22 items, categorized into four dimensions: Idea Generation (5 items), Idea Promotion (6 items), Idea Realization (6 items), and Idea Sustainability (5 items). Each dimension is exemplified by specific items, such as “Specifying which elementary improvements can be implemented at work.” for Idea Generation, “Convincing others of the importance of a newly developed idea or solution” for Idea Promotion, “Monitoring the progress during the process of putting ideas in practice” for Idea Realization, and “Communicating explicitly the returns of the implemented idea outside the team” for Idea Sustainability. In the current study, reliability analyses revealed satisfactory internal consistency for each dimension, with Cronbach’s Alpha (α) and McDonald’s Omega (ω) scores as follows: Idea Generation (α = 0.783; ω = 0.752), Idea Promotion (α = 0.851; ω = 0.852), Idea Realization (α = 0.906; ω = 0.906), and Idea Sustainability (α = 0.874; ω = 0.878). The total scale had satisfactory reliability indices (α = 0.949; ω = 0.955), while CFA showed a satisfactory fit of the four-factor model [TLI = 0.992; CFI = 0.993; GFI = 0.986; NNFI = 0.992; RMSEA = 0.041 (0.032–0.049); SRMR = 0.067].

Burnout was assessed using a short version of the Oldenburg Burnout Inventory (OLBI), adapted on a sample of Greek teachers (Gkontelos et al., 2023a). The OLBI includes 9 items divided into two dimensions: Exhaustion (4 items) and Disengagement (5 items). A sample item for Exhaustion is “After working, I have enough energy for my leisure activities.”, while for Disengagement, it is “I find my work to be a positive challenge”. Reliability analyses for the present study demonstrated satisfactory internal consistency for both dimensions, with Cronbach’s Alpha (α) and McDonald’s Omega (ω) scores of α = 0.804; ω = 0.826 for Exhaustion and α = 0.723; ω = 0.731 for Disengagement. The total scale had satisfactory reliability indices (α = 0.830; ω = 0.853), while CFA showed a satisfactory fit of the two-factor model [TLI = 0.969; CFI = 0.978; GFI = 0.981; NNFI = 0.969; RMSEA = 0.055 (0.033–0.077); SRMR = 0.076].

The Greek version of the Teacher Irrational Belief Scale (G-TIBS), which consists of 21 items, was validated and used in previous research (Gkontelos et al., 2021). The G-TIBS is structured into four dimensions: Self-Downing (6 items), Authoritarianism (6 items), Demands for Justice (5 items), and Low Frustration Tolerance (4 items). Sample items include “I can’t stand being criticized or thought badly of when I haven’t finished something or done it properly” (Self-Downing), “Students should always be respectful, considerate and behave well” (Authoritarianism), “Teachers should be consulted about decisions, which affect their teaching, and it’s really unfair when they are not” (Demands for Justice), and “I find it too hard to balance my work and home demands” (Low Frustration Tolerance). Reliability analyses revealed satisfactory internal consistency for each dimension, with Cronbach’s Alpha (α) and McDonald’s Omega (ω) scores as follows: Self-Downing (α = 0.736; ω = 0.720), Authoritarianism (α = 0.722; ω = 0.708), Demands for Justice (α = 0.764; ω = 0.775), and Low Frustration Tolerance (α = 0.741; ω = 0.745). The total scale had satisfactory reliability indices (α = 0.860; ω = 0.875), while CFA showed a satisfactory fit of the four-factor model [TLI =.972; CFI = 0.975; GFI = 0.966; NNFI = 0.972; RMSEA = 0.037 (0.027–0.046); SRMR = 0.065].

3. Results

Table 1. Descriptive statistics, correlation matrix, and internal consistency measures.

Variable	SD	AUTH	DJ	LFT	IG	IP	IR	IS	DIS	EXH
SD	—
AUTH	0.546 ***	—
DJ	0.451 ***	0.327 ***	—
LFT	0.484 **	0.419 ***	0.383 ***	—
IG	0.070	−0.086	0.211 ***	0.009	—
IP	0.089	−0.017	0.222 ***	0.012	0.795 ***	—
IR	0.088	−0.037	0.241 ***	0.038	0.792 ***	0.823 ***	—
IS	−0.011	−0.042	0.134 *	−0.006	0.628 ***	0.725 ***	0.651 ***	—
DIS	0.209 ***	0.251 ***	0.078	0.458 ***	−0.281 ***	−0.252 ***	−0.225 ***	−0.227 ***	—
EXH	0.327 ***	0.236 ***	0.205 ***	0.516 ***	−0.060	−0.113 *	−0.059	−0.102	0.561 ***	—
Mean	2.696	2.143	3.921	2.435	4.039	3.734	3.993	3.322	1.815	2.636
SD	0.738	0.614	0.665	0.847	0.620	0.721	0.698	0.971	0.644	0.895
Alpha, α	0.736	0.722	0.764	0.741	0.783	0.851	0.906	0.874	0.804	0.723
Omega, ω	0.720	0.708	0.775	0.745	0.752	0.852	0.906	0.878	0.826	0.731

Note: *** p < 0.001, ** p < 0.01, * p < 0.05. SD: self-downing; AUTH: authoritarianism; DJ: demands for justice; LFT: low frustration tolerance; IG: idea generation; IP: idea promotion; IR: idea realization; IS: idea sustainability; DIS: disengagement; EXH: exhaustion.

presents the descriptive statistics (means and standard deviations) of the distributions of each dimension, namely IWB (idea generation, idea promotion, idea realization, idea sustainability), IB (self-downing, authoritarianism, demands for justice, low frustration tolerance) and burnout (exhaustion, disengagement). Furthermore, a correlation matrix is presented to examine the preliminary relationships among the aforementioned variables.

Statistically significant correlations were detected among the IB subscales. Specifically, SD was correlated with AUTH (r = 0.546, p < 0.001, 95% CI [0.466; 0.617]), DJ (r = 0.451, p < 0.001, 95% CI [0.362; 0.533]), and LFT (r = 0.484, p < 0.001, 95% CI [0.398; 0.562]). AUTH was correlated with DJ (r = 0.327, p < 0.001, 95% CI [0.228; 0.419]), and LFT (r = 0.419, p < 0.001, 95% CI [0.327; 0.504]). DJ was correlated with LFT (r = 0.383, p < 0.001, 95% CI [0.288; 0.471]). Similarly, statistically positive significant correlations were observed among IWB dimensions. IG was correlated with IP (r = 0.795, p < 0.001, 95% CI [0.752; 0.831]), IR (r = 0.792, p < 0.001, 95% CI [0.748; 0.828]), and IS (r = 0.628, p < 0.001, 95% CI [0.558; 0.688]). IP was correlated with IR (r = 0.823, p < 0.001, 95% CI [0.785; 0.855]), and IS (r = 0.725, p < 0.001, 95% CI [0.670; 0.772]). IR was correlated with IS (r = 0.651, p < 0.001, 95% CI [0.585; 0.709]). A strong positive and statistically significant relationship was detected between burnout’s subscales. DIS was correlated with EXH (r = 0.561, p < 0.001, 95% CI [0.483; 0.630]). Among the subscales of the latent variables, alternative correlations were observed. SD was not correlated statistically significant with IG (r = 0.070, p = 0.197, 95% CI [−0.037; 0.176]), IP (r = 0.089, p = 0.104, 95% CI [−0.018; 0.194]), IR (r = 0.088, p = 0.106, 95% CI [−0.019; 0.193]), and IS (r = −0.011, p = 0.838, 95% CI [−0.118; 0.096]). AUTH was not correlated statistically significant with IG (r = −0.086, p = 0.116, 95% CI [−0.191; 0.021]), IP (r = −0.017, p = 0.755, 95% CI [−0.124; 0.090]), IR (r = −0.037, p = 0.496, 95% CI [−0.143; 0.070]), and IS (r = −0.042, p = 0.447, 95% CI [−0.148; 0.066]). DJ was correlated statistically significant with IG (r = 0.211, p < 0.001, 95% CI [.137; 0.311]), IP (r = 0.222, p < 0.001, 95% CI [0.118; 0.321]), IR (r = 0.241, p < 0.001, 95% CI [0.137; 0.339]), and IS (r = 0.134, p = 0.014, 95% CI [0.027; 0.237]). LFT was not correlated statistically significant with IG (r = 0.009, p = 0.867, 95% CI [−0.098; 0.116]), IP (r = 0.012, p = 0.827, 95% CI [−0.095; 0.119]), IR (r = 0.038, p = 0.484, 95% CI [−0.069; 0.145]), and IS (r = −0.006, p = 0.918, 95% CI [−0.112; 0.101]). DIS and EXH revealed different correlations among the other variables. To be more precise, DIS were correlated statistically significantly with SD (r = 0.209, p < 0.001, 95% CI [0.105; 0.309]), AUTH (r = 0.251, p < 0.001, 95% CI [0.148; 0.348]). LFT (r = 0.458, p < 0.001, 95% CI [0.369; 0.539]), IG (r = −0.281, p < 0.001, 95% CI [−0.377; −0.180]), IP (r = −0.252, p < 0.001, 95% CI [−0.349; −0.149]), IR (r = −0.225, p < 0.001, 95% CI [−0.324; −0.121]), IS (r = −0.227, p < 0.001, 95% CI [−0.326; −0.123]), while no statistically significant correlation was detected with DJ (r = 0.078, p = 0.151, 95% CI [−0.029; 0.184]). On the contrary, EXH was correlated significantly only with IB subscales, SD (r = 0.327, p < 0.001, 95% CI [0.228; 0.419]). AUTH (r = 0.236, p < 0.001, 95% CI [0.132; 0.334]), DJ (r = 0.205, p < 0.001, 95% CI [0.100; 0.305]) and LFT (r = 0.516, p < 0.001, 95% CI [0.433; 0.590]). EXH showed no statistically significant correlations with most IWB dimensions; however, a small but statistically significant negative correlation was observed with idea promotion (r = –0.113, p = 0.039, 95% CI [–0.217; –0.006]). Its correlations with idea generation (r = –0.060, p = 0.269, 95% CI [–0.166; 0.047]), idea realization (r = –0.059, p = 0.283, 95% CI [–0.164; 0.049]), and idea sustainability (r = –0.102, p = 0.062, 95% CI [–0.206; 0.005]) were not statistically significant.

The aforementioned results indicated that the two burnout’s dimensions, EXH and DIS, affect IWB differently. Prior to conducting the regression analysis, collinearity diagnostics were performed. The obtained values (Tolerance = 0.616; 0.793; VIF = 1.261; 1.623) indicated the absence of multicollinearity concerns, allowing the analysis to proceed. A multiple linear regression [F(3, 333) = 13.462, p < 0.001] was applied for further confirmation. The extracted model (

Table 2. Multiple linear regression model predicting IWB.

Model		Unstandardized	Std. Error	Standardized	t	p
M0	(Intercept)	3.772	0.037		103.136	<0.001
M1	(Intercept)	3.743	0.186		20.118	<0.001
	DIS	−0.362	0.066	−0.347	−5.497	<0.001
	EXH	0.011	0.049	0.014	0.215	0.830
	IB	0.235	0.072	0.191	3.283	0.001

Note: DIS: disengagement; EXH: exhaustion; IB: irrational beliefs.

) contained an examination of the impact that IB and Burnout have on IWB, while the Adjusted R-Square was 32.9%. For the predicted variable (IWB), the three independent variables performed differently. To be precise, DIS was identified as a negative associate factor (b = −0.362; t = −5.497; p < 0.001), while conversely IB were a positive associate factor (b = 0.235; t = 3.283; p = 0.001). EXH did not depict statistically significant results (b = 0.011 t = 0.215; p = 0.830).

A hierarchical cluster analysis was conducted to explore whether participants could be grouped into distinct profiles based on their scores in DIS, EXH, and IB. This method was selected because it does not require a priori assumptions regarding the number of clusters and is suitable for exploratory purposes when the underlying group structure is unknown. The analysis yielded an explained variance of R-squared = 0.275, AIC = 761.13, BIC = 818.43, and a Silhouette coefficient of 0.650, which indicated a satisfactory clustering solution in terms of internal consistency. A five-cluster solution was retained as the most interpretable and theoretically meaningful. The distribution of participants across the clusters was as follows: Cluster 1 (N = 78, 23.2%), Cluster 2 (N = 51, 15.1%), Cluster 3 (N = 93, 27.6%), Cluster 4 (N = 29, 8.6%), and Cluster 5 (N = 86, 25.5%). The cluster profiles are summarized in

Table 3. Description of Cluster Profiles based on DIS, EXH, and IB.

	Cluster 1	Cluster 2	Cluster 3	Cluster 4	Cluster 5
N (%)	78 (23.2%)	51 (15.1%)	93 (27.6%)	29 (8.6%)	86 (25.5%)
DIS	High	Medium	Medium	Medium	Medium
EXH	Medium	Medium	Low	High	Medium
IB	Medium	Low	Medium	Medium	High
Silhouette	0.641	0.710	0.718	0.874	0.461

Note: High: Values > 0.5, Medium: Values −0.5; +0.5, Low: Values < −0.5. DIS: disengagement; EXH: exhaustion; IB: irrational beliefs.

. Figure 1 illustrates the mean scores of each cluster on burnout’s dimensions and IB, highlighting distinct psychological profiles.

To further examine the external validity of the identified clusters, a one-way analysis of variance was conducted with IWB as the dependent variable. This step contributed to testing whether the five clusters, which were derived from DIS, EXH, and IB, also exhibited meaningful differences in a theoretically relevant outcome variable. The rationale was that if the clusters represent distinct psychological profiles, they should vary systematically in their levels of IWB. The ANOVA revealed a statistically significant effect of cluster membership on IWB [F(4,332) = 5.322, p < 0.001, η² = 0.06]. Post hoc comparisons (Tukey HSD) were conducted following a non-significant Levene’s test for homogeneity of variances (F = 1.341; p = 0.254), indicating that the assumption of equal variances was met. Descriptive statistics indicated that Cluster 1 had a mean IWB of 3.530 (SD = 0.727), Cluster 2 had a mean IWB of 3.660 (SD = 0.684), Cluster 3 had a mean IWB of 3.849 (SD = 0.658), Cluster 4 had a mean IWB of 3.790 (SD = 0.588), whereas Cluster 5 had IWB mean score (M = 3.969, SD = 0.581), suggesting potential differences in IWB across the identified psychological profiles. The results showed that Cluster 1 scored significantly lower in IWB compared to Cluster 3 (mean difference = −0.320; p = 0.014) and to Cluster 5 (mean difference = −0.440; p < 0.001), whereas differences between the remaining clusters were not statistically significant (see

Table 4. Post-hoc Comparisons (Tukey HSD) among Cluster Profiles.

Clusters	Clusters	Mean Difference	Std. Error	t	d	p
1	2	−0.131	0.118	−1.110	−0.200	0.801
	3	−0.320	0.101	−3.181	−0.488	0.014 *
	4	−0.260	0.142	−1.828	−0.398	0.359
	5	−0.440	0.102	−4.294	−0.671	<0.001 ***
2	3	−0.189	0.114	−1.656	−0.289	0.463
	4	−0.129	0.152	−0.850	−0.198	0.915
	5	−0.309	0.116	−2.668	−0.471	0.061
3	4	0.059	0.139	0.427	0.091	0.993
	5	−0.120	0.098	−1.223	−0.183	0.738
4	5	−0.179	0.141	−1.275	−0.274	0.707

Note: *** p < 0.001, * p < 0.05.

).

To further illustrate the characteristics of the identified clusters,

Table 5. Cluster Characteristics Across Psychological Variables and IWB.

	Cluster 1	Cluster 2	Cluster 3	Cluster 4	Cluster 5
N (%)	78 (23.2%)	51 (15.1%)	93 (27.6%)	29 (8.6%)	86 (25.5%)
DIS	High	Medium	Medium	Medium	Medium
EXH	Medium	Medium	Low	High	Medium
IB	Medium	Low	Medium	Medium	High
IWB	Low	Medium	High	High	High
Descriptive label	Disengaged-Low Innovators	Resilient-Balanced Innovators	Adaptive Innovators	Strained but Innovative Innovators	Belief-Driven Innovators

Note: DIS: disengagement; EXH: exhaustion; IB: irrational beliefs; IWB: innovative work behavior.

presents a qualitative summary of each cluster in terms of DIS, EXH, IB, and IWB. It is important to note that the clusters were derived solely from DIS, EXH, and IB, and IWB was not included in the clustering procedure. Despite this, the clusters exhibit distinct patterns in IWB, with some clusters showing relatively higher levels of IWB and others showing lower levels. This qualitative differentiation indicates that the clusters capture meaningful differences in participants’ psychological profiles, which extend beyond the input variables and reflect variation in a theoretically relevant outcome.

Taken together, the above results indicate that the identified clusters not only capture meaningful psychological profiles but also differ in their levels of IWB, providing a robust basis for further interpretation.

4. Discussion

The present study aimed to examine the relationship between IB, Burnout, and IWB among early childhood education teachers. On this purpose, both variable-centered and person-centered approaches were employed. Findings seem to partly support the initial hypotheses. Specifically, disengagement emerged as the only burnout dimension that significantly and negatively predicted IWB, whereas exhaustion did not show a direct effect, supporting the research choice for their distinct examination (Zhai et al., 2025). Interestingly, IB displayed a positive association with IWB, contrary to the initial assumption that higher IB would hinder innovation. In addition, the cluster analysis revealed five distinct teacher profiles, highlighting the heterogeneity of psychological patterns within the sample and suggesting that the interaction of burnout and IB with IWB may be more complex than linear models imply.

The findings of the present investigation indicate that disengagement constitutes a robust negative predictor of IWB, whereas exhaustion did not demonstrate a statistically significant direct effect. This distinction bears considerable theoretical import within the conceptual parameters of the Job Demands–Resources model (Bakker & Demerouti, 2014), which theorizes that burnout emerges from a protracted disequilibrium between occupational demands and available resources. Exhaustion encapsulates the emotional, cognitive, and somatic strain engendered by sustained exposure to work-related stressors; nonetheless, it may predominantly remain internalized, manifesting as attenuated energy levels without necessarily precluding engagement in innovative initiatives. Conversely, disengagement encompasses both a withdrawal from professional tasks and a deterioration of occupational commitment, thereby directly undermining the propensity to engage in creative problem-solving and the implementation of novel ideas-fundamental facets of IWB. This interpretation aligns with extant literature indicating that psychological detachment and cynicism exert more deleterious effects on proactive and IWB than fatigue per se (Demerouti et al., 2010; Koch & Adler, 2018).

Moreover, disengagement can be conceptualized as a manifestation of motivational depletion, wherein educators exhibit diminished investment in their professional role and increasingly adopt negative orientations toward work content (Saraiva & Nogueiro, 2025). Such attitudinal shifts are likely to suppress intrinsic motivation and constrain cognitive flexibility, both of which are pivotal for fostering innovation in educational contexts. Disengaged teachers may construe innovative endeavors as futile or irrelevant, thereby attenuating their contribution to organizational improvement and transformative change (Gurgel, 2023; Page et al., 2024). The absence of a direct effect of exhaustion on IWB implies that, although physical and emotional fatigue compromise overall well-being, substantial impairment of IWB may necessitate an additional threshold of motivational disengagement. Collectively, these results underscore the imperative of addressing disengagement as a critical impediment to teachers’ IWB, carrying profound theoretical and practical implications for educational management and policy (Fei & Tien, 2024; Gkontelos et al., 2023b; Hassan et al., 2024).

Intriguingly, the present study revealed a positive association between IB and IWB, a finding that may appear counterintuitive given the conventional negative characterization of IB. This result can be interpreted through the prism of the dual nature of cognitive schemas within professional contexts (Santarpia et al., 2023). Although IB are typically regarded as maladaptive, under specific circumstances they may function as motivational catalysts, prompting educators to challenge entrenched practices and explore alternative solutions to surmount perceived personal or professional limitations (Bernard, 2016; Ellis, 2005). IB’s two-fold role can lead to an increase correlating to higher performance levels, a factor for personal growth (Gkontelos et al., 2025).

Particularly, certain facets of IB, such as self-downing or low frustration tolerance, may induce cognitive dissonance that motivates problem-solving and proactive engagement, as teachers endeavor to reconcile their performance with internalized standards (Bakshi, 2023). This observation aligns with the theoretical proposition that psychological tension can paradoxically foster creativity and innovation when channeled constructively (Amabile, 1996; Potočnik et al., 2022). Within educational settings, teachers experiencing moderate levels of IB may demonstrate heightened involvement in idea generation, advocacy, and implementation, perceiving innovative practices as instruments to address perceived deficits or institutional constraints (Ugwuanyi, 2022).

Furthermore, the positive linkage between IB and IWB highlights the nuanced interplay of cognitive–emotional dynamics in the workplace. It suggests that the effects of IB are not universally detrimental but are contingent upon contextual variables, individual resilience, and professional agency (Žeželj & Lazarević, 2019). From a psychological standpoint, this relationship may be underpinned by mechanisms such as cognitive dissonance and coping regulation, whereby the tension created by irrational cognitions activates self-reflective processes and adaptive effort. In this way, resilience and constructive coping may transform potentially maladaptive beliefs into motivational resources. Educators equipped with effective coping strategies and supported by conducive work environments may translate the motivational tension generated by IB into concrete innovative outcomes, underscoring a nonlinear relationship between cognitive biases and professional performance (Katsiki et al., 2017). Unlike earlier studies that predominantly reported negative associations between IB and professional functioning (Gundogan, 2025), the present findings suggest a more differentiated pattern. In particular, exhaustion appeared more strongly linked to diminished innovative engagement than disengagement, implying that emotional depletion may inhibit creative output, whereas cognitive withdrawal might coexist with compensatory motivational processes driven by IB. Our findings corroborate contemporary perspectives positing that personal beliefs, even those conventionally labeled as irrational, may serve adaptive functions when oriented toward problem-focused, goal-directed action (Gkontelos et al., 2023b; Turner et al., 2024).

The hierarchical cluster analysis identified five distinct teacher profiles based on levels of disengagement, exhaustion, and ΙΒ, thereby highlighting the heterogeneity among educators and elucidating the interplay of psychological traits in shaping IWB. Notably, the clusters exhibited significant differences in IWB despite its exclusion from the clustering procedure, thereby supporting the external validity of these profiles. For interpretive and practical purposes, these profiles were assigned descriptive labels reflecting their dominant characteristics: Disengaged-Low Innovators (Cluster 1), Resilient-Balanced Innovators (Cluster 2), Adaptive Innovators (Cluster 3), Strained but Innovative Innovators (Cluster 4), and Belief-Driven Innovators (Cluster 5).

Teachers encompassed in Cluster 1 (Disengaged-Low Innovators) were characterized by high disengagement coupled with moderate exhaustion and IB. This group displayed the lowest levels of IWB, suggesting that disengagement constitutes a substantial barrier to innovation. The diminished emotional and motivational investment in work appears to suppress the generation, promotion, and implementation of novel ideas, consistent with prior findings linking disengagement to reduced organizational contribution (Afi, 2025; Bakker & Demerouti, 2007; Xanthopoulou et al., 2007). Additionally, teachers in Cluster 2 (Resilient-Balanced Innovators) were characterized by moderate disengagement and exhaustion but low IB. Teachers in this cluster exhibited intermediate IWB levels, indicating that lower IB may facilitate adaptive functioning even under work-related stress. This profile underscores the moderating influence of cognitive schemas in mitigating the adverse effects of burnout on IWB (Khan & Shamsi, 2021; Ogakwu et al., 2024). Moreover, Cluster 3 (Adaptive Innovators) and Cluster 4 (Strained but Innovative Innovators), further demonstrated that exhaustion alone does not reliably predict diminished IWB. Cluster 3, with low exhaustion, displayed high IWB, suggesting that moderate disengagement combined with minimal strain allows teachers to sustain creative and innovative engagement despite the presence of IB (Chaves-Montero et al., 2025). Cluster 4, despite high exhaustion, also manifested relatively elevated IWB, indicating that certain educators can maintain innovation under strain, potentially through intrinsic motivation or effective coping mechanisms (Nwoko et al., 2024; Zargar et al., 2025). Lastly, teachers in Cluster 5 (Belief-Driven Innovators) were distinguished by high levels of IB alongside moderate disengagement and exhaustion. Remarkably, this group attained the highest IWB scores. This finding corroborates the earlier regression results, highlighting that IB may, under specific circumstances, function as a catalyst for innovation (Santarpia et al., 2023). Teachers in this cluster appear to leverage the motivational tension generated by their cognitive biases to engage in creative problem-solving and proactive innovative practices (Liedtka, 2015; Xu et al., 2024).

To conclude, the cluster analysis underscores the complex and nonlinear interactions among psychological variables in predicting IWB. Disengagement emerges as the most consistent negative predictor, whereas exhaustion and IB exert context-dependent effects. These findings emphasize the utility of a person-centered approach, as variable-centered analyses may obscure subgroups of educators who, despite moderate stress or the presence of IB, demonstrate high innovative potential. Recognizing these nuanced profiles can inform targeted interventions, enabling school leaders to foster innovation by addressing disengagement, bolstering resilience to exhaustion, and strategically harnessing the motivational dimensions of cognitive patterns such as IB.

4.1. Practical Implications

The findings of the present study carry significant implications for educational practice and teacher development. The results indicate that teachers’ levels of disengagement, exhaustion, and IB interact in complex and nuanced ways to shape their IWB. Of particular note, high disengagement emerged as a substantial impediment to innovation, underscoring the critical importance of fostering engagement and intrinsic motivation. Educational institutions can support teachers by offering meaningful opportunities for professional autonomy, promoting active participation in decision-making processes, and cultivating collegial collaboration to enhance connectedness and organizational commitment (Kousina & Voudouris, 2023; Xafakos & Malafantis, 2025).

Simultaneously, moderate levels of exhaustion do not appear to inhibit IWB, highlighting the potential for cognitive and emotional challenges to be harnessed productively. Interventions such as stress management training, workload optimization, and reflective professional learning can facilitate the transformation of strain into creative problem-solving and adaptive instructional practices (Córdova et al., 2023; Pesqueira et al., 2025). Moreover, IB, though frequently deemed maladaptive, may function as motivational drivers when teachers are guided to recognize and strategically manage these cognitive patterns (Ekwueme et al., 2023; Stamovlasis & Vaiopoulou, 2017). Professional development programs emphasizing metacognitive awareness and structured reflection can enable educators to channel rigid beliefs into goal-directed strategies that foster innovation (Koulianou et al., 2025; Nian, 2020; Zsigmond et al., 2025).

This study underscores the necessity of tailoring support to the psychological profiles of teachers rather than adopting a one-size-fits-all approach. Establishing a school climate that encourages risk-taking, collaboration, and psychological safety allows educators with diverse cognitive and emotional profiles to flourish (Vasiou et al., 2025). By addressing engagement, well-being, and cognitive flexibility holistically, school leaders can maximize teachers’ capacity for IWB, thereby contributing to organizational development, instructional quality, and enhanced student learning outcomes (Lakkala et al., 2021; Polatcan et al., 2023; Vasiou et al., 2024). To be more precise, differentiated strategies can be proposed for each teacher’s profile to optimize IWB. Disengaged-Low Innovators may benefit from interventions aimed at enhancing engagement and promoting professional agency. Resilient-Balanced teachers could be supported through leadership opportunities and peer-mentoring roles that consolidate adaptive functioning. Adaptive Innovators may require institutional resources and recognition to sustain creative output without overextension. Strained but Innovative educators might profit from structured well-being and stress-management programs to maintain innovation while mitigating risk of burnout. Finally, Belief-Driven Innovators could be guided through metacognitive and reflective practices that channel strong personal convictions into goal-directed, innovative action. Teachers exhibiting high levels of IB or high disengagement may benefit most from these targeted interventions, including professional development programs, stress-management initiatives, and structured reflective practices, which can help convert cognitive–emotional challenges into constructive, goal-directed engagement.

4.2. Limitations

Despite the valuable insights yielded by the present study, several limitations warrant consideration. First, cross-sectional design constrains the ability to draw causal inferences regarding relationships among IB, burnout, and IWB. Longitudinal and experimental research is therefore necessary to examine the temporal dynamics of these constructs and to elucidate potential causal pathways. Second, the sample was restricted to early childhood educators in Greece, predominantly female, which may limit the generalizability of the findings to other educational levels, cultural contexts, or more gender-balanced populations. Third, exclusive reliance on self-report measures introduces susceptibility to social desirability bias and common-method variance, potentially inflating observed associations. Moreover, the use of hierarchical cluster analysis can cause some methodological limitations. The results may be sensitive to the choice of distance measures and linkage methods, potentially affecting the identified cluster structure. Finally, while person-centered analyses provided valuable insights into distinct psychological profiles, other pertinent individual and organizational factors, such as leadership, school climate, peer support, and professional development, were not examined and could further clarify the mechanisms underlying IWB in educational settings.

Building on these limitations, future research should adopt longitudinal and experimental designs to explore the causal effects and developmental trajectories of IB, burnout, and IWB. Expanding the participant pool to encompass diverse teacher populations and cross-cultural contexts would enhance external validity. Incorporating additional individual and contextual variables, alongside objective behavioral indicators, could yield a more comprehensive understanding of the determinants and inhibitors of innovation in schools. Moreover, advanced person-centered methodologies, such as latent profile analysis, computational intelligence techniques or growth mixture modeling, may facilitate a more precise delineation of subgroups and enable the investigation of nonlinear patterns (Gkevrou et al., 2025; Gkevrou & Stamovlasis, 2025; Mastrothanasis et al., 2024; Shaikh et al., 2024). Finally, intervention studies targeting IB and burnout are recommended to assess their efficacy in promoting teachers’ IWB, thereby informing both theoretical frameworks and practical applications in educational settings.

5. Conclusions

This study underscores the intricate interrelationships among teachers’ IB, burnout, and IWB, demonstrating that professional innovation is influenced by an interplay of personal and contextual factors. Disengagement emerged as a salient impediment to innovation, whereas certain IB did not exert uniformly negative effects, highlighting the nuanced and multifaceted impact of cognitive–emotional dynamics on teacher performance. By delineating diverse psychological profiles, and assigning descriptive labels to them (e.g., Disengaged, Low Innovators, Resilient-Balanced, Adaptive Innovators, Strained but Innovative, Belief-Driven Innovators), the findings accentuate the necessity of tailored interventions that address specific constellations of burnout and belief patterns, rather than adopting undifferentiated intervention strategies. Acknowledging these labeled profiles enhances the practical utility of this study by providing educational leaders with concrete targets for differentiated support and policy planning. These insights carry significant practical implications for educational leaders aiming to foster teacher well-being, resilience, and a culture of innovation, thereby enhancing both individual professional growth and the overall effectiveness and adaptability of schools in contemporary educational contexts.