A Multi-Layered Socio-Ecological Framework for Investigating Teacher Well-Being: Key Predictors and Protective Factors

Abstract

Understanding the factors that influence teacher well-being is crucial as it significantly affects students, teachers, schools, and the sustainability of the education system, especially during prolonged emergencies. This study contributes to the field by empirically testing a conceptual model of teacher well-being in emergency contexts, specifically addressing the COVID-19 school closures with a sample of over 19,600 teachers from Kazakhstan through an online survey design. Utilising a multidimensional socio-ecological framework that considers individual, school and home, community, and national factors, this study identifies key predictors of teacher self-reported well-being. Individual-level predictors explained 9.3% of the variation in physical well-being (F = 118, p < 0.001, R² = 0.093) and 4.5% in psychological well-being (F = 72.2, p < 0.001, R² = 0.045). In contrast, school- and home-level predictors demonstrated significantly greater explanatory power, accounting for 21.9% (F = 128, p < 0.001, R² = 0.219) and 15.5% (F = 89.5, p < 0.001, R² = 0.155) of the variation in physical and psychological well-being, respectively. Community-level predictors explained 12.8% of the variation in physical well-being (F = 191, p < 0.001, R² = 0.128) and 10.2% in psychological well-being (F = 324, p < 0.001, R² = 0.102), while national-level predictors accounted for much smaller proportions: 0.67% for physical well-being (F = 21.8, p < 0.001, R² = 0.0067) and 1.4% for psychological well-being (F = 83.589, p < 0.001, R² = 0.014). These findings highlight the significant influence of home and school, as well as community-level predictors, on teacher well-being during emergency contexts, suggesting that interventions targeting these areas may be particularly effective in supporting teacher well-being. The findings reveal that while Kazakhstani teachers reported poor physical well-being, they generally had a more positive assessment of their psychological well-being. Vulnerable groups included women, older teachers, non-Kazakh teachers, and those with higher education levels, as well as teachers in Russian medium and mixed-medium schools, all of whom reported lower physical and psychological well-being. Additional risk factors identified were a lack of student engagement, difficult relationships with parents, a directive leadership style, family conflicts, and inadequate resources at home and school. Conversely, protective factors such as teacher autonomy, collegiality, networking opportunities, and self-efficacy emerged as significant contributors to well-being. These findings reveal a complex interplay between cultural factors and subjective perceptions of well-being. This study emphasises the critical role of these predictors in both emergency and non-emergency contexts, underscoring the urgent need for targeted policies and programmes that sustainably support and enhance teacher well-being holistically. This approach will promote Sustainable Development Goal (SDG) 3 (well-being) and ensure access to equitable quality education (SDG 4) for all learners, ultimately contributing to the overall resilience of educational systems.

1. Introduction

Contemporary neoliberal times expose teachers to unprecedented stress [1], as performativity management and target-driven accountability systems exert constant pressure on their well-being [2,3]. The global trend towards one-size-fits-all reforms aimed at enhancing efficiency and student outcomes has not only undermined teachers’ professional self-worth but has also increased their workload, resulting in greater vulnerability and burdens [4,5]. In light of these challenges, it becomes vital to recognise the significant implications of teacher well-being as past studies have linked teacher well-being to enhanced student cognitive, social, and emotional development and enriched classroom interactions [6,7]. Conversely, teacher stress is associated with decreased performance, leading to lowered student achievements, heightened burnout rates, and increased attrition [2,8,9,10]. This detrimental cycle threatens the sustainability of education systems and undermines efforts to promote well-being (Sustainable Development Goal (SDG) 3) and ensure access to quality education (SDG 4) for all learners [11].

During emergencies like the COVID-19 pandemic, teachers encounter heightened demands as existing stressors intensify and new challenges arise [8,12], significantly impacting their well-being (SDG 3). Unlike previous localised crises, such as violent conflicts or natural disasters, the COVID-19 pandemic resulted in unprecedented school closures globally, disrupting access to quality education (SDG 4) [13]. While technology played a crucial role in mitigating educational disruptions, millions were excluded from remote learning, revealing the potential adverse effects of excessive technology use on learning outcomes and well-being [14,15].

The pandemic posed unique challenges for teachers, including unclear policy guidance, limited systemic support, and technology gaps in resource-scarce and rural areas [16]. Teachers faced difficulties with digital pedagogy, declining student engagement, increased workloads, and isolation from professional networks [8,12,16,17]. While positive effects of professional development focused on personal growth and positive psychology have been reported [11], global evidence shows a rise in teachers’ depression, stress, anxiety, and burnout, adversely affecting their well-being [18,19,20,21] and increasing the risk of teacher turnover [8,22,23]. Given the multitude of factors affecting teacher well-being, understanding localised barriers and enablers is essential for developing effective post-pandemic policies and recommendations that support teachers.

This study investigates teacher well-being during COVID-19 school closures through an online survey completed by 19,635 teachers from all 17 regional education authorities in Kazakhstan. Unlike previous research that typically focuses on limited variables at the individual or school level, this study adapts and empirically tests Falk et al.’s comprehensive framework for understanding teacher well-being in low-resource and emergency contexts [24]. This modified framework examines a wide range of protective factors and stressors across individual, school, home, community, and national levels. As discussed in Section 3, where we review the global literature on teacher well-being during the COVID-19 emergency, there is a notable lack of research in the Global South, particularly in Kazakhstan, a middle-income country with distinct educational, cultural, and societal characteristics. These gaps in existing studies underscore the significance of the current research as it enhances our understanding of teacher well-being in resource-scarce contexts and aligns with the Sustainable Development Goals (SDGs). Furthermore, the extensive coverage across Kazakhstani regions and the large sample size makes this the first comprehensive study on teacher well-being in Kazakhstan.

2. Conceptualising Teacher Well-Being: A Socio-Ecological Framework

Socio-ecological perspectives originating from the seminal research of Bronfenbrenner [25] highlight the necessity of addressing broader environmental and organisational factors rather than focusing solely on individual-level factors to explore well-being [26]. Building on Bronfenbrenner’s [23] framework, Falk et al. [23] developed a conceptual model for teacher well-being in emergency and conflict settings, which posits that teacher well-being is shaped by an interplay of contextual influences, encompassing teachers’ emotions, attitudes, and perceptions related to their professional roles. The current study further adapted Falk et al.’s framework to investigate teacher well-being during the pandemic.

Teacher well-being is viewed along two dimensions, physical and psychological, which corresponds to job stress and burnout in Falk et al.’s framework [24]. Job stress encompasses negative emotional and physical responses to one’s work, while burnout is a product of stress leading to feelings of exhaustion, ineffectiveness, powerlessness, alienation, and isolation. Most studies on teacher well-being in the COVID-19 context have centred on job stress and burnout [27,28,29,30,31,32].

Other constructs in Falk et al.’s [24] framework include teacher self-efficacy, job satisfaction, and socio-emotional competence. Teacher self-efficacy refers to a teacher’s confidence in their ability to steer and achieve desired results in student behaviour, learning, and accomplishments [33]. Job satisfaction reflects the sense of contentment and fulfilment a teacher derives from their profession [33]. Social–emotional competence denotes a teacher’s adeptness in comprehending, expressing, and regulating their own emotions proficiently, along with their capability to perceive and react to others’ emotions in a socially appropriate manner [33]. All three constructs significantly influence teacher well-being [30,34].

The socio-ecological framework views teacher well-being as a dynamic state influenced by four environmental levels: individual, school and home, community, and national contexts. This layered approach emphasises interactions within and between these levels, highlighting factors that can either support well-being (protective factors) or elevate stress (risk factors), with effects differing across various groups and settings. A systematic literature review indicates that teacher well-being is influenced at all four levels [26].

2.1. Level 1 Individual: Teacher

The teacher is at the centre of the socio-ecological framework, as illustrated in the innermost circle of Figure 1. At the individual level, the framework examines how various teacher characteristics influence well-being, including socio-demographics such as gender [27], age [35], ethnicity [29], level of education [36], teaching level (primary, secondary, and high school) [37], teaching experience [38], and self-efficacy [37]. Additionally, factors such as family status, number of children, and caring for children with special educational needs also play a significant role in shaping teacher well-being [20,29,39,40].

2.2. Level 2 Contextual: School and Home

In Figure 1, the second circle from the centre represents teachers’ direct engagements within their microsystem, which expanded from the school to the home during the COVID-19 emergency. This extension emphasises the need to incorporate additional elements into Falk et al.’s [24] framework, underscoring the importance of teacher–student, teacher–parent, teacher–family, and peer relationships, alongside school leadership factors, in supporting teacher well-being [26,29,34,41,42]. Teachers’ well-being is significantly shaped by their interactions with students, colleagues, and administrators. The interconnection between their personal and professional lives means that stress or satisfaction in one domain can affect the other [43]. Positive relationships within the classroom and among colleagues foster emotional support and resilience, while collaboration facilitated by school managers enhances well-being [44].

Additionally, personal relationships outside of work contribute to overall well-being by providing crucial emotional support [43]. Additionally, essential resources, including school-provided laptops and home amenities such as designated teaching spaces and high-speed Internet access, play a crucial role [15]. The allocation of these resources is influenced by school characteristics, including location (rural vs. urban), language of instruction, and type of school (private, elite, or public) [13,17], all integrated into our expanded framework.

2.3. Level 3 Community

Depicted as the third circle in Figure 1, this level includes factors in the mesosystem such as respect, recognition, responsibility, and duty. Respect and recognition pertain to the acknowledgement of teachers’ efforts and contributions from various sources, including the community, media, and especially parents [45]. The responsibility and duty aspect involves balancing professional obligations with personal life as teachers navigate work commitments alongside family responsibilities and other social roles [32,42,46]. External societal pressures, such as public scrutiny of teaching or unrealistic community expectations from teachers, can further exacerbate teacher stress [32,47]. Additionally, this level encompasses the exposure to COVID-19 and its physical and emotional impacts within the community [48].

2.4. Level 4: National Environments

The final tier addresses factors within the macrosystem, encompassing the governance structure and policy landscapes at the national and devolved educational authority levels that directly influence schools and teachers. This includes aspects such as teacher management styles, compensation structures, professional development programmes, and teacher autonomy [24,44].

Limited teacher autonomy can exacerbate teacher burnout and stress levels, especially in emergency contexts [28,49]. Moreover, emergencies may cause disruptions in salaries or delays in payments, further heightening stress [24,28]. While crises demand an increased emphasis on professional development to adapt to new circumstances, inadequate training during these periods can diminish teachers’ self-efficacy and elevate stress levels [26,50].

While studies utilising the socio-ecological framework enhance the understanding of teacher well-being, research on the interactions between ecological layers and their effects remains limited [26]. Hartcher et al.’s review found only one study among fifty-two that examined these interactions [51], but it had a small sample size (n = 226) and was conducted in a Western context (the UK), measuring teacher well-being at only two levels: individual and contextual, which included predictors at the school and home levels. The current study aims to address this gap by exploring teacher well-being across ecological layers with a significantly larger sample. We now provide an overview of the literature on teacher well-being during the COVID-19 emergency, as viewed through the lens of our multi-layered conceptual framework.

3. Teacher Well-Being During the COVID-19 Emergency: A Global Review

Teaching in emergency settings can significantly impact teachers’ well-being, with various factors at the individual, school, home, community, and national levels playing a significant role.

3.1. Individual

3.1.1. Gender

Gender stands out as a predominant factor influencing group differences in teachers’ well-being across various studies [20,27,31,35,36,37,39,40,48,52,53,54]. Studies consistently show that female teachers tend to experience poorer well-being compared to their male counterparts, although some studies do not find a significant statistical impact of gender on teachers’ well-being [31,36]. Notably, research highlights that more female teachers report issues like insomnia and dizziness, indicating lower physical well-being [54]. A study involving Chilean teachers suggests that being a female teacher is linked to higher job and life satisfaction, yet female teachers also report more significant levels of depression and anxiety [39].

3.1.2. Age and Teaching Experience

Age and teaching experience emerged as significant predictors of teacher well-being, albeit with mixed effects. Some studies suggested that younger teachers experienced poorer well-being compared to older educators [35,37], while others found better well-being associated with younger age [27,31,35,39,40,52]. Interestingly, Bianchi et al. noted a correlation between age and teaching experience [18]. Surprisingly, well-being tended to decline with age, contradicting expectations that older teachers with more experience would excel in adapting to teaching challenges, such as those presented during the pandemic. This unexpected trend may stem from computer anxiety and reduced self-efficacy in utilising information and communication technology (ICT) among older educators [17,27].

3.1.3. Family Status and Children

Family status had a varied effect on teachers’ well-being. Married teachers tended to have lower well-being compared to single teachers [40]. However, contrasting findings by Fermin suggested that married postgraduate teachers were more proficient in expressing emotions than their single counterparts [31]. Sigursteinsdottir and Rafnsdottir highlighted that single teachers were prone to experiencing more stress compared to married or cohabitating teachers [20]. Some studies did not find a significant correlation between marital status and teachers’ well-being [31].

Children and the number of children served as negative predictors of teachers’ occupational well-being since balancing teaching responsibilities with supporting their children’s online learning places added demands on teachers [42]. The shift to online education proved especially challenging for teachers with childcare duties, leading to a decrease in their overall well-being [29,40,55].

3.1.4. Ethnicity

Ethnicity is a less explored factor in predicting teacher well-being, with only a limited number of studies investigating its effect [29,35,53]. Hutchison et al. [35] did not find any significant effect of ethnicity, while Baker et al. [29] found that Black teachers in the US exhibited better mental health compared to White teachers despite facing more challenges. They attributed this outcome to the superior coping mechanisms employed by Black teachers.

3.1.5. Educational Qualifications

The impact of educational qualifications on teacher well-being during the pandemic differs. While some studies did not find academic qualifications as a significant predictor of teacher well-being [31,35,36,56], Ma et al. revealed that teachers with college degrees reported higher anxiety levels than those with Bachelor’s or Master’s degrees [38].

3.1.6. Self-Efficacy

Self-efficacy plays a significant role in teacher well-being [38,45], with teachers’ confidence in managing student behaviour online showing a negative correlation with stress and anxiety levels [40,57]. In general, teachers with higher self-efficacy tend to have better well-being. Nonetheless, various studies highlight that the shift to online teaching presented obstacles to teachers’ confidence in their instructional capabilities [15].

3.2. School- and Home-Level Factors

3.2.1. School Characteristics

Teacher well-being may vary based on school characteristics. One study indicated that rural teachers experienced less anxiety compared to their urban counterparts [40]. Additionally, teachers in private subsidised schools were more likely to exhibit poor mental health compared with those in private schools. Differences in well-being were also observed based on school grade levels, with secondary school teachers reporting lower fear of COVID compared to primary and preschool teachers [36], (p. 12).

3.2.2. Teachers’ Access to Digital Resources

Access to digital resources had a profound impact on teachers’ well-being. In certain instances, the absence of school support necessitated teachers to personally invest in tools such as laptops and Internet connections, while limited home space exacerbated their challenges [46,58], leading to heightened anxiety and stress levels related to resource sharing [42].

3.2.3. Student-Related Issues

A range of student-related factors significantly influenced teachers’ well-being during the COVID-19 emergency. Issues such as access to resources, technical hurdles, student health, and family complexities affected student engagement in online learning, subsequently impacting teachers’ mental health [57]. The additional workload of supporting student well-being and maintaining continuous communication via digital means affected teacher well-being [41,46,56,59].

Additionally, teachers faced several challenges linked to students’ well-being that affected their own mental health. Concerns about learning loss, academic support, and the cognitive advancement of students were prevalent [41,59]. Balancing academic expectations with student well-being induced tensions among teachers [29], leading to feelings of anxiety and guilt regarding online classroom management, student engagement, and motivation [12,46,59,60]. Teachers grappled with disruptive student behaviour, including harassment and disengagement, which often had negative effects on teacher morale [60]. Conversely, establishing direct connections with students and having more opportunities for individualised work emerged as crucial factors that boosted teacher morale and well-being [12].

3.2.4. Relationship with Parents

Interactions with parents during emergency remote teaching increased teachers’ workload. Some studies report that enhanced parental involvement improved relationships and supported student learning, empowering teachers [12,42,61]. However, other studies reported that teachers found frequent communication with parents for academic, technical, and emotional support to be challenging [58,60]. Additionally, maintaining contact with students’ families and providing essential supplies added to teachers’ workload and stress [42].

3.2.5. Peer Relationship

Collegial relationships are crucial protective factors for teacher well-being, especially when peers are knowledgeable in online teaching and learning [46]. Several studies highlight the importance of peer support across professional, personal, emotional, and practical realms, leading to enhanced life satisfaction, work engagement, and overall teacher well-being [12,18,29,60]. The collaborative efforts among teachers in remote teaching involved teamwork, task sharing, and displays of care, creating a more supportive and productive work atmosphere [41]. While this collaborative approach helped save time and energy for teaching activities, some teachers expressed missing their colleagues during this period [62].

3.2.6. Teacher and School Leadership Relationships

Teacher well-being is significantly influenced by teacher and school leadership relationships, which can act as both protective and risk factors. Positive leadership styles, such as collegial and autonomy-supportive leadership, tend to safeguard teacher well-being [12,28,57,62]. School leaders’ practical and emotional support also played a crucial role in supporting teachers [12]. Conversely, autonomy-thwarting leadership has been linked to emotional exhaustion and is considered a risk factor [63]. Authoritative leadership styles and a bureaucratic environment can negatively impact teachers’ mental health due to increased control, pressures, and a lack of recognition of teacher workload and well-being [59]. While some teachers view control positively as caring supervision [34], inadequate support from school leaders during times of increased workload and stress, compounded by a lack of parental support, can contribute to teacher attrition [61].

3.2.7. Teachers’ Family Relationships

Teachers’ family interactions can act as both protective and risk factors for their well-being. Research by Jakubowski and Sitko-Dominik [52] suggested that relationship break-ups increased anxiety levels, while supportive relationships decreased stress, highlighting the impact of social and family connections on teacher well-being. Family support is crucial for teachers, aiding in balancing work–life boundaries and managing expectations [42,46]. Studies indicate that teachers who receive emotional and household support from their families or friends experience higher well-being and life satisfaction [64]. During lockdowns, increased family time was recognised as beneficial for teacher well-being, emphasising the importance of staying connected with loved ones [29].

3.3. Community-Level Factors

3.3.1. Respect and Appreciation

Acknowledging and valuing teachers is crucial for their well-being. Research during the COVID-19 emergency highlighted how teachers appreciated recognition but felt disrespected [8,42,59,62,65]. Support and respect from the community positively impacted teachers’ morale, motivation, and sense of belonging amidst isolation [8,62,65]. Teachers appreciated personal connections, recognition, and supportive leadership, emphasising the importance of respect and recognition in fostering job satisfaction and retention [8,62].

3.3.2. Gender Norms and Child/Family Care

Juggling duties in emergency remote teaching posed significant challenges as work and home boundaries blurred, leading to conflicting roles for teachers [42,55]. This was especially pronounced for parent–teachers and female educators, exacerbating the struggles. The increased workload, coupled with childcare responsibilities during remote learning, demanded more time and attention from teachers [42]. As a result, teachers often had to make difficult choices, prioritising work over personal obligations, which led to feelings of anxiety, frustration, and guilt [42]. Additionally, time constraints related to family and childcare created pressure that could strain family relationships, ultimately impacting teacher well-being [56,66].

3.3.3. Exposure to and Fear of COVID-19 Infection

Factors such as COVID-19 exposure, illness, and fear further negatively impacted teacher well-being, underscoring how community circumstances directly influence individuals [58]. Exposure to COVID-19, illness among colleagues and students, cases of death, anxiety about the negative consequences of COVID-19, and fear of catching the illness were among the most important factors which negatively impacted teacher well-being [48].

3.4. National-Level Factors

Factors beyond school boundaries significantly shape the well-being of teachers. Research in various cultural contexts underscores the importance of workplace security and autonomy for teacher well-being, particularly in times of emergency [67,68]. Teacher well-being is adversely affected in education systems characterised by high accountability and limited autonomy [68]. The absence of suitable compensation amidst heightened workloads during emergency remote teaching can intensify stress and financial pressures, impacting teacher engagement and mental health [8,52]. Prioritising national-level initiatives to address teacher compensation and financial support is crucial for safeguarding teacher well-being during challenging times.

Teacher preparation for remote teaching appeared as a significant factor during the COVID-19 emergency. Numerous studies have highlighted the lack of training for remote education, leading to heightened teacher stress, increased workload, diminished teaching satisfaction, and limited ICT self-efficacy [30,59,68]. Studies demonstrate a shift towards technology training in low-resource settings [46], while better-equipped settings highlighted the significance of teacher training in positive psychology, stress management, mindfulness, and resilience to support teacher well-being during emergency remote teaching [11,69]. These findings demonstrate that teacher well-being is context-specific, and protection and risk for teacher well-being depend on national systems that regulate the setting in which teachers work.

We conclude this review with several key observations. First, although there is a substantial body of research on teacher well-being during the COVID-19 pandemic, significant gaps remain in studies focused on non-Western settings, particularly in developing countries, where the challenges faced by schools and teachers could be profoundly different, providing valuable alternative insights. Second, the variability in the relationship between teacher well-being and socio-demographic factors—such as gender, age, and marital status—underscores the need for context-specific studies rather than broad generalisations. Additionally, while most studies have employed pre-existing validated instruments to measure well-being, few have adapted culturally relevant questionnaires tailored to local contexts or specific emergencies based on prior research, preliminary qualitative studies, or instrument piloting [54,55,56]. Finally, no studies have examined the different levels of influence—such as individual, home, school, and community—within a single study. There is also a scarcity of research focusing on rural teachers or comparing the well-being of rural and urban teachers [40,53,70]. Furthermore, ethnicity and cultural dynamics have received limited attention despite their potential effects on coping mechanisms and systemic disparities. In summary, this literature review underscores that teacher well-being during emergencies is highly context-specific, shaped by a complex interplay of personal, institutional, and societal factors.

4. Teacher Well-Being in Kazakhstan

4.1. Teacher Management, Incentives, and Education

Teacher management in Kazakhstan is profoundly shaped by a Soviet-era, communistic, and collectivistic approach, where teachers primarily serve as implementers of educational policies established by higher authorities. Despite ongoing reforms aimed at fostering critical thinking, creativity, and outcome-based education [71], school leadership continues to prioritise technical, bureaucratic, and top-down decisions that fail to adequately support teachers [72].

Teacher compensation is a critical factor affecting teacher well-being. After the collapse of the Soviet Union in the 1990s, inadequate pay and salary delays led to significant teacher attrition [73]. This decline in prestige resulted in fewer talented candidates enrolling in teacher education programmes, with many motivated primarily by state scholarships rather than a genuine desire to teach.

However, the situation began to improve after the 2010s, driven by educational reforms and increased investment [74]. The Kazakhstani government launched the State Programme on Educational and Science Development for 2016–2020 to enhance the prestige of the teaching profession and address low applicant motivation by regulating salary increases and stipends for teacher candidates [75]. In 2019, the government also approved the Law on the Status of a Teacher to further strengthen the occupational prestige and esteem of teachers.

This initiative marked a significant step towards revitalising the teaching profession and attracting more qualified candidates through salary increases and other provisions. The salary rises were linked to professional development and the successful completion of nationwide programmes, which included three system-level courses on outcome-based education, ensuring both salary increases and advancements in teacher qualifications. Before the COVID-19 pandemic, all professional training concentrated on educational reform. According to the Teaching and Learning International Survey, while many teachers participated in ICT training, there was still a strong demand for additional training [76].

4.2. Teachers and the COVID-19 Emergency

During the pandemic, the Kazakhstani government ensured the timely payment of salaries to teachers and provided additional compensation for adjusted teaching hours due to changing class sizes [77]. Kazakhstan faced one of the longest school closures globally, which began abruptly on 16 March 2020, when students were sent home for an emergency holiday [16]. Online classes commenced on 5 April 2020.

During this period, teachers encountered significant challenges, including a highly centralised curriculum, poor Internet connectivity, an urban–rural divide in digital resources, and gender disparities in work–life balance [49,58,70,78]. Existing studies indicate that these disruptions widened educational inequalities and negatively affected the well-being of parents, students, teachers, and school leaders, particularly women [17,79,80].

During the pandemic, decision making remained top-down [79], which impacted pedagogy, assessment, and curriculum [17]. Although school leaders claimed to grant teachers autonomy in curricular decision making and the selection of technological tools and platforms [81], teachers themselves felt this autonomy was limited [58].

As a result, teachers faced significant pressure from parents to ensure student success and from school administration to maintain educational quality [17]. This lack of support and the accompanying uncertainty negatively affected teachers’ well-being [49]. Furthermore, while urban schools typically had more resources, Amirova et al. found no statistically significant relationship between school location and Internet speed [78].

In contrast, rural teachers reported feeling more freedom and support, which contributed to lower stress levels compared to their urban counterparts. This positive experience is attributed to more supportive management and greater autonomy in rural schools [80]. However, rural school teachers, particularly older ones, faced significant challenges in effectively using online technologies [17].

5. Methodology

5.1. Aims and Research Questions

This study aimed to examine the impact of the COVID-19 emergency on the well-being of Kazakhstani school teachers and to identify the factors influencing their self-reported well-being across four environmental levels: individual, school and home, community, and national contexts. The following research questions guided this study:

• How are the individual-, school- and home-, community-, and national-level factors derived from the conceptual framework distributed among the participating teachers?
• What do teachers’ self-reported well-being scores indicate about the effects of COVID-19-induced school closures on their overall well-being?
• What individual-, school- and home-, community-, and national-level factors significantly predict teachers’ well-being scores?

5.2. Study Design

This study utilised an online descriptive cross-sectional survey design, which is effective for understanding current conditions, identifying relationships, informing policy, and exploring diverse populations at a single point in time [81,82]. Additionally, this method was both cost-effective and efficient for collecting the experiences of a large number of school teachers across a vast country comparable in size to Western Europe.

5.3. The Instrument

The development of the survey was informed by a qualitative study which aimed to understand the experiences of Kazakhstani teachers with distance and online schooling [58], as well as the emerging literature on teachers’ well-being during the COVID-19 pandemic.

The survey comprised five sections. Section 1 covered socio-demographic questions. Section 2 asked about the support teachers received from the school to manage the shift to online education. Section 3 explored teachers’ experiences and views on school closures, while Section 4 covered teachers’ perspectives on the impact of school closures on their students. Section 5 sought to understand the impact of the pandemic on teachers’ well-being. The dependent variable of interest is teachers’ well-being during the COVID-19 emergency. The independent variables included a range of factors across the four levels of teacher environmental contexts.

5.3.1. Quality Issues

Several steps were taken to ensure the validity of the survey. Insights from the initial qualitative study with Kazakhstani teachers helped align the survey with participants’ contexts, enhancing its content validity and cultural relevance [83].

The survey was initially created in English and subsequently translated into Kazakh and Russian, the two primary languages in Kazakhstan. The translation process was managed collaboratively by research team members proficient in all three languages. Adaptive translation was conducted to preserve conceptual meaning, followed by a “backward” translation to identify discrepancies [84]. Furthermore, valuable input was gathered through online consultations with two experts in the field, as mentioned in the Acknowledgment Section. The survey was piloted with practising teachers whose feedback led to refinements in phrasing and additional response categories to some items.

The reliability of the scales also provides insights into their validity. The well-being scale reported in this paper includes two sub-scales, with Cronbach’s Alpha indicating acceptable internal consistency: physical well-being (α = 0.810) and psychological well-being (α = 0.734).

Nevertheless, as discussed in Section 7.2, this study has limitations related to the sampling method and the subjective measurement of well-being. Therefore, the findings should be interpreted with caution.

5.3.2. Administration and Recruitment

The survey was developed in the Qualtrics platform. Data collection took place between the end of May and October 2021. After receiving ethical approval from the authors’ Institutional Research Ethics Committee, permission for data collection was obtained from regional education authorities, who facilitated connections with school leaders. Teachers were sent the survey link via their group email shared by school leaders (Appendix A). The link included a description of the study and information about voluntary participation. Ethical considerations regarding data collection are addressed in Section 5.6.

5.4. Sample and Participants

Volunteer non-probability sampling was employed, characterised as a form of convenience sampling where participation mainly relies on the respondents due to the non-individualised nature of the invitations [85]. We invited all Kazakhstani school teachers to participate in this study, and those who chose to take part were self-selecting. This sampling method was chosen due to the well-documented challenges associated with using probability sampling in educational research in Kazakhstan [86], as well as our interest in capturing insights from a diverse group of school teachers representing all regions of the country. A total of 19,635 teachers responded. However, responses to the well-being items of the survey ranged from 9459 to 14,459 teachers.

5.5. Data Analysis

Survey data were exported from Qualtrics into Jamovi 2.3.28 for statistical analyses. To address Research Question 1, univariate analysis was conducted to obtain descriptive statistics and assess the normality of the distribution of both independent and dependent variables, following Tabachnik and Fidell’s recommendations [87]. Research Question 2 was examined using factor analysis to determine the factor structure of the teacher well-being scale [88]. For Research Question 3, multiple standard regression analysis was employed to identify individual-, contextual-, community-, and national-level factors that might explain variations in school teachers’ well-being scores and their contributions to this variation. Prior to this analysis, a bivariate analysis was performed to explore the relationships between pairs of variables, with the choice of test based on the level of measurement and the distribution’s normality. These tests, which include One-way ANOVA, Mann–Whitney U test, Spearman Rho, and Kruskal–Wallis test, are detailed in Appendix B.

5.6. Ethics

This study adhered to the ethical principles of voluntary informed consent and participant anonymity. Informed consent was included in the survey, allowing only those who selected ‘Yes’ to proceed. The survey was entirely anonymous, with no identifying information collected, ensuring that education authorities, school leaders, and researchers could not ascertain participants’ identities. This anonymity provided a secure platform for participants to express their views and share their experiences. Respondents had the flexibility to complete the survey at their convenience. They were only required to answer the consent question, and they had the option to skip any questions if they preferred.

6. Results

6.1. Distribution of Dependent Variables Across the Four Environmental Levels

6.1.1. Individual-Level Variables

Female teachers comprised 88% of the sample, exceeding the national statistic of 81% [89] (

Table 1. Frequency distribution of variables at the individual level.

Characteristics	Categories	Percentage
Gender (n = 14,456)	Female	88.0
	Male	12.0
Age (n = 14,459), Mean = 39.8
Experience (n = 13,372), Mean = 15.9
Family status (n = 14,446)	Single	13.7
	Married/couple	76.4
	Single parent	9.9
Number of children (n = 14,438)	No children	16.6
	One	14.0
	Two	26.0
	Three	23.5
	Four	13.5
	Five	4.8
	More than five	1.7
Monthly income (n = 14,378)	Below national average	44.3
	National average	20.2
	Above national average	35.6
Ethnicity (n = 14,459)	Kazakh	82.6
	Russian	9.4
	Others	8.0
Educational level (n = 14,139)	College	7.0
	Bachelor	84.5
	Graduate/postgraduate	8.5
Teaching level (n = 13,650)	Primary	27.7
	High/secondary	52.9
	Both	19.4
Self-efficacy: self-belief in increasing proficiency in delivering online lessons (n = 12,725)	Yes	75.3
	No	24.7

). The average age of the respondents was 39.80 years, which aligns with national statistics [90]. Most respondents were married (76.4%) and parents of two or more children (69.4%). Additionally, most teachers reported their income as either below average (44.3%) or above average (35.6%), with 35.6% indicating that their income was similar to the average national income. In terms of ethnicity, Kazakhs constituted 82.6% of the participants, while Russians made up 9.4%.

Most participants held higher education degrees, with only 7% having vocational or college degrees. About 52.9% taught at the secondary/high school level, 27.7% taught primary grades, and around 20% taught both levels. On average, teachers had 16 years of teaching experience (M = 15.9).

The data on teachers’ self-efficacy regarding online teaching reveal a generally positive outlook, with a majority (77.8%) of teachers feeling confident or very confident about teaching effectively online (Figure 2). Additionally, the majority (75.3%) expressed that their proficiency in delivering online lessons has increased over time.

6.1.2. School- and Home-Level Variables

The variables at the school and home levels included school characteristics; relationships with students, parents, and peers; interactions with school leadership; family relations; and available resources (

Table 2. Percentage distribution of variables at school and home level.

Dimensions	Variables	Categories	Percentage
School characteristics	School location (n = 14,421)	Urban	50.0
		Semi-urban	9.0
		Rural	41.0
	Medium of instruction used (n = 14,303)	Kazakh	51.8
		Russian	16.5
		Mixed	31.6
Relationships with parents	Pressure to justify student assessment results to parents (n = 13,422)	No	19.8
		Yes	80.2
	Lower parental perception of teacher effort due to remote teaching (n = 13,376)	No	41.4
		Yes	58.6
Peer relations	Improvement in teacher collegiality and collaboration (n = 12,542)	No	45.2
		Yes	54.8
Relationship with school leadership	Perceived surveillance by the school administration (n = 13,021)	No	59.7
		Yes	40.3
Workload	I have to work long hours to do my online teaching. (n = 13,650)	No	32.2
		Yes	67.8
Family relationships	Frequent conflict resolution over space and resources among children (n = 11,194)	No	52.0
		Yes	48.0
	In times of crisis, we can turn to each other for support (n = 12,365)	No	12.5
		Yes	87.5

).

Half of the teachers came from urban schools (50%), while the other half were from semi-urban (9%) and rural schools (41%). Slightly over half worked in schools with a Kazakh medium of instruction (51.8%), while others were in Russian (16.5%) and mixed-language schools (31.6%).

The majority of teachers faced challenges with digital resources (Figure 3). Specifically, 84.2% reported poor Internet connectivity as challenging or extremely challenging for blended teaching, while 68.7% struggled to find a quiet space for the same purpose.

Teachers also faced challenges related to their students (Figure 4). Notably, nine out of ten teachers reported that not being able to see their students online due to turned-off cameras was challenging or extremely challenging. They also found the lack of insight into student learning caused by academic integrity issues, as well as low student motivation, to be significant obstacles.

Teachers’ relationships with parents revealed that most teachers (80.2%) felt pressured to justify student assessment results to parents (Table 2). Additionally, perceptions regarding parental views on teacher effort during remote teaching were mixed, with 58.6% believing that parents think they are putting in less effort.

Regarding workload, the majority of teachers (67.8%) felt they had worked extra hours due to remote teaching. On a positive note, slightly over half (54.8%) believed that teacher collegiality and collaboration improved during this period (Table 2). Additionally, most teachers (59.7%) disagreed with the notion that they felt under surveillance by the school administration.

Teachers’ family relationships also showed a positive trend, with the majority (87.5%) feeling they could rely on one another during times of crisis. However, about half (48%) reported struggling to resolve frequent conflicts among their children regarding space and resources.

6.1.3. Community-Level Variables

Community-level variables included items related to responsibility and work-life balance, as well as respect and recognition and exposure to COVID-19 (

Table 3. Percentage distribution of variables at the community level.

Dimension	Variables	Categories	Percentage
Responsibility and duty	Household chore expectations during work hours (n = 12,734)	No	58.2
		Yes	41.8
	Family care responsibilities (n = 12,649)	Not challenging	27.1
		Challenging	66.6
		Extremely challenging	6.4
	Remote teaching and flexible distribution of work and domestic responsibilities (n = 13,720)	No	59.0
		Yes	41.0
	Feeling overwhelmed by parenting responsibilities (n = 11,746)	No	44.0
		Yes	56.0
Respect and recognition	Increased parental respect post-homeschooling experience (n = 11,748)	No	37.5
		Yes	62.5
Exposure to infection	Exposure to COVID (n = 14,410)	No	77.8
		Yes	22.2

).

Around 42% of teachers reported being expected by their family members to contribute to household chores during work hours, with a higher percentage of women (43%) expressing this expectation compared to their male peers (33%) in the sample (p < 0.001, Chi-Square = 51.8, df = 1). However, the strength of this relationship was relatively weak (Cramer’s V = 0.0638).

The majority of teachers (73%) reported finding family care responsibilities challenging or extremely challenging, with female teachers (74%) experiencing this more than their male (66%) counterparts (p < 0.001, Chi-Square = 53.2, df = 2). However, the strength of this relationship was weak (Cramer’s V = 0.0649).

Only 41% of teachers agreed that remote teaching provided them with the flexibility to better balance their work and domestic responsibilities. A higher percentage of male teachers (47.5%) agreed compared to their female counterparts (42.2%) (p < 0.001, Chi-Square = 53.2, df = 2). However, the strength of this relationship was weak (Cramer’s V = 0.0272).

The majority of teachers (56%) reported feeling overwhelmed by parenting responsibilities, with female teachers (56.5%) experiencing this more than their male counterparts (52%) (p < 0.005, Chi-Square = 31.8, df = 1). However, the strength of this relationship was negligible (Cramer’s V = 0.0482).

The majority of teachers (62.5%) felt that remote teaching led to increased respect from parents. In contrast, only 22.9% of teachers reported that they or their family members were exposed to COVID-19.

6.1.4. National-Level Variables

National-level predictors included teachers’ autonomy in choosing online platforms and revising the curriculum, the presence of a top-down management approach, and the availability of networking and professional development opportunities in online teaching. (

Table 4. Percentage distribution of variables at the national level.

	Variables	Categories	Percentage
Professional autonomy	Freedom to choose online platforms (n = 14,459)	No	29.2
		Yes	70.8
	Freedom to revise curriculum (n = 14,459)	No	82.4
		Yes	17.6
	Top-down approach used in decision making (n = 14,459)	No	91.9
		Yes	8.1
Networking	Opportunities to network with teachers (n = 14,459)	No	67.7
		Yes	32.3
Professional development	Provision of training in online teaching (n = 14,442)	No	33.0
		Yes	67.0

).

The vast majority of teachers (70.8%) reported that their school allowed them the freedom to choose an online platform. However, teacher autonomy in revising the curriculum for remote settings was limited, with only 17.6% of teachers agreeing they had this autonomy. Rather surprisingly, only 8.1% of teachers agreed that the school management adopted a top-down approach to managing schools remotely. While networking opportunities were available to only 32.3% of teachers, a significant majority (67.0%) reported receiving professional development opportunities in online teaching.

6.2. Teachers’ Well-Being Scale

Teachers’ well-being was measured through seven items using a rating scale from strongly agree (1) to strongly disagree (4). Principal Axis Factoring with Varimax rotation and a listwise option for handling missing values confirmed two factors—physical well-being. Bartlett’s test of sphericity and Kaiser–Meyer–Olkin (KMO) indicated that the seven items were appropriate for factor analysis (KMO = 0.759; Barlett’s test of sphericity (χ²(21) = 26,823.865, p < 0.001)). Parallel analysis was also conducted to verify the number of factors [87,90]. The eigenvalues in both the Principal Component Analysis and Parallel Analysis were compared and confirmed, retaining two factors (

Table 5. Comparison of eigenvalues from PA * and PCA **.

Random Eigenvalues from PA	Data Eigenvalues PCA	Decision
1.047860	3.194	Accept
1.033653	1.655	Accept

* Parallel Analysis; ** Principal Component Analysis.

).

In total, these two factors explained 63% of the variance. The physical Well-being factor accounted for approximately 40% of the variance, while the psychological well-being factor accounted for over 23% of the variance. The internal consistency reliability estimate for both the physical well-being (α = 0.810;

Table 6. Physical well-being factor.

	Mean	Std. Deviation	Loading	Factor Score
I get frequent headaches because of excessive screen time.	2.13	0.808	0.873	0.537
I feel my eyesight is declining because of excessive screen time	2.06	0.834	0.803	0.316
I have developed back problems because of sitting too much.	2.23	0.792	0.631	0.174
Total Mean	2.142
N	14,459
Cronbach’s Alpha	0.810
Factor eigenvalue	2.800
Total variance explained %	40.007
Cumulative variance explained %	40.007

) factor and the psychological well-being (α = 0.734;

Table 7. Psychological well-being factor.

	Mean	Std. Deviation	Loading	Factor Score
I feel depressed.	3.68	0.948	0.765	0.411
I become sad too often.	3.46	1.064	0.649	0.278
The pandemic is making me feel stressed.	3.27	1.149	0.598	0.231
I do not socialise outside my immediate household.	3.46	1.064	0.556	0.186
Total Mean	3.467
N	14,459
Cronbach’s Alpha	0.734
Factor eigenvalue	1.635
Total variance explained %	23.352
Cumulative variance explained %	63.359

) showed acceptable values. We saved scores for each factor using regression methods. The physical and psychological well-being factors were positively correlated (r = 0.328).

Overall, participating teachers tended to agree (M = 2.142, Table 6) that they faced issues related to physical well-being, with eyesight being the most significant concern (M = 2.07). Self-reported psychological well-being among Kazakhstani teachers during school closures was better than their physical well-being. Teachers generally disagreed with the notion of poor psychological well-being (M = 3.467, Table 7), expressing the strongest disagreement with “I feel depressed” (M = 3.68).

6.3. Predictors of Kazakhstani Teachers’ Well-Being

Bivariate tests were initially conducted to examine the relationships between predictors and teachers’ physical and psychological well-being based on the level of measurement and the normality of the distribution (Appendix B). The criteria for inclusion were theoretical relevance, linearity of relationships, and significant correlations between predictors and well-being factors [78]. A cut-off point for correlation was set at 0.005 due to the large sample size (n = 14,459).

All individual-level predictor variables significantly related to physical and psychological well-being were included in the multiple linear regression model, except for teaching experience, which exhibited multicollinearity with the age variable (Appendix C).

6.3.1. Individual-Level Predictors of Physical and Psychological Well-Being

Individual-level predictors provided a significant model for predicting both the physical and psychological well-being of teachers. Two models were run for each well-being measure (

Table 8. Multiple linear regression analysis of individual-level predictors and teacher well-being.

Variable	Physical		Psychological
Predictors	R2	β	R2	β
Model 1	0.097 ***		0.046 ***
Gender. Male (0) vs. female (1)		−0.357 ***		−0.196 ***
Age		−0.010 ***		−0.004 ***
Family status. Single (0) vs. married/couple (1)		−0.039		0.035
Family status. Single (0) vs. single parent (1)		−0.079		−0.047
Number of children		0.020 *		0.016 *
Family monthly income. Below (0) vs. average (1)		−0.082 ***
Family monthly income. Below (0) vs. above average (1)		−0.132 ***
Ethnicity Kazakh (0) vs. Russian (1)		−0.315 ***		−0.071 *
Ethnicity. Kazakh (0) vs. other (1)		−0.245 ***		−0.115 ***
Education qualification. College (0) vs. Bachelor (1)		−0.146 ***
Education qualification. College (0) vs. postgraduate (1)		−0.111 *
Level of teaching. Primary (0) vs. secondary/high		0.011		0.002
Level of teaching. Primary (0) vs. both		0.024		0.024
Self-confidence in teaching effectively online. Not confident (0) vs. little confident (1)		−0.009		0.109 *
Self-confidence in teaching effectively online. Not confident (0) vs. quite confident (1)		0.161 ***		0.305 ***
Self-confidence in teaching effectively online. Not confident (0) vs. very confident (1)		0.363 ***		0.483 ***
Improvement in online teaching proficiency. No (0) vs. yes (1)		−0.024
Model 2	0.093 ***		0.045 ***
Gender. Male (0) vs. female (1)		−0.350 ***		−0.204 ***
Age		−0.011 ***		−0.005 ***
Number of children		0.017 *		0.021 ***
Family monthly income. Below (0) vs. average (1)		−0.092 ***
Family monthly income. Below (0) vs. above average (1)		−0.126 ***
Ethnicity. Kazakh (0) vs. Russian (1)		−0.331 ***		−0.070 **
Ethnicity. Kazakh (0) vs. other (1)		−0.255 ***		−0.123 ***
Education qualification. College (0) vs. Bachelor (1)		−0.137 ***
Education qualification. College (0) vs. postgraduate (1)		−0.124 **
Self-confidence in teaching effectively online. Not confident (0) vs. little confident (1)		−0.010		0.096 *
Self-confidence in teaching effectively online. Not confident (0) vs. quite confident (1)		0.140 ***		0.292 ***
Self-confidence in teaching effectively online. Not confident (0) vs. very confident (1)		0.343 ***		0.467 ***

Note. * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001.

). Models for physical and psychological well-being met all multicollinearity assumptions. The tolerance statistics for physical well-being ranged from 0.953 to 0.996, while those for psychological well-being ranged from 0.955 to 0.997. Additionally, the variance inflation factor (VIF) values for both measures were below two, indicating low or no multicollinearity. Model 2 for physical well-being (F = 118, p < 0.001) accounted for 9.3% of the variation, while Model 2 for psychological well-being (F = 72.2, p < 0.001) explained 4.5% of the variation.

Teachers’ self-efficacy in teaching effectively online positively predicted both physical and psychological well-being. Teachers who reported high confidence in online teaching indicated improved well-being. Those who are “quite confident” have higher physical well-being (B = 0.140, p < 0.001), and the strongest positive association is for “very confident” (B = 0.343, p < 0.001). For psychological well-being, teachers expressing varying levels of confidence—quite confident (B = 0.292, p < 0.001) and very confident (B = 0.467, p < 0.001)—also reported better outcomes compared to those who were not confident at all.

Gender, age, and educational qualifications negatively predicted teacher well-being. Female teachers reported worse physical (B = −0.350, p < 0.001) and psychological well-being (B = −0.204, p < 0.001) compared to their male counterparts.

Physical well-being decreases as age increases, with a small negative standardised effect (B = −0.011, p < 0.001) and with similar effects on psychological well-being (B = −0.005, p < 0.001). Likewise, higher levels of education negatively predicted teachers’ physical well-being. Teachers with a Bachelor’s degree (B = −0.137, p < 0.001) and a postgraduate degree (B = −0.124, p < 0.001) reported worse physical well-being compared to those with lower or vocational levels of education.

Ethnicity emerged as a significant predictor. Ethnic Russian teachers report significantly lower physical well-being compared to Kazakhs (B = −0.0.331, p < 0.001). Similarly, other ethnicities report lower well-being compared to Kazakhs (B = −0.0255, p < 0.001).

The number of children was a positive predictor of teachers’ self-reported well-being, indicating that a greater number of children was associated with increased physical (B = 0.017, p = 0.025) and psychological well-being (B = 0.020, p = 0.001).

For physical well-being, teachers with average (B = −0.092, p < 0.001) and above-average (B = −0.126, p < 0.001) income reported significantly lower levels compared to those with below-average income.

6.3.2. School- and Home-Level Predictors

School- and home-level predictors of teacher well-being during emergency remote teaching significantly predicted both physical and psychological well-being. Two models were run for each well-being measure (

Table 9. Multiple linear regression analysis of school- and home-level predictors and teacher well-being.

Variable	Physical		Psychological
Predictors	R2	β	R2	β
Model 1	0.223 ***		0.156 ***
Medium of instruction. Kazakh (0) vs. Russian (1)		−0.310 ***		−0.130 ***
Medium of instruction. Kazakh (0) vs. mixed (1)		−0.114 ***		−0.095 ***
Not being able to see students’ reactions. Not at all challenging (0) vs. challenging (1)		−0.160 ***		−0.024
Not being able to see students’ reactions. Not at all challenging (0) vs. extremely challenging (1)		−0.282 ***		−0.070
Not knowing students’ learning because of collusion/plagiarism. Not at all challenging (0) vs. challenging (1)		−0.111 **		−0.019
Not knowing students’ learning because of collusion/plagiarism. Not at all challenging (0) vs. extremely challenging (1)		−0.330 ***		−0.104 *
Low student motivation. Not at all challenging (0) vs. challenging (1)		−0.137 ***		−0.055
Low student motivation. Not at all challenging (0) vs. extremely challenging (1)		−0.244 ***		−0.103 *
I have to explain and justify to parents the assessment of their child’s work. No (0) vs. yes (1)		−0.227 ***		−0.141 ***
Because I am teaching from home, parents think I am not working hard enough. No (0) vs. yes (1)		−0.248 ***		−0.281 ***
Teacher collegiality and collaboration improved during online teaching. No (0) vs. yes (1)		0.091 ***		0.080 ***
I feel under constant surveillance by the school principal/deputy director(s). No (0) vs. yes (1)		−0.029		−0.286 ***
I have to constantly resolve arguments between my children over space and resources. No (0) vs. yes (1)		−0.090 ***		−0.251 ***
In times of crisis, we can turn to each other for support. No (0) vs. yes (1)		−0.167 ***		0.123 ***
Poor Internet connectivity. Not at all challenging (0) vs. challenging (1)		−0.182 ***		−0.023
Poor Internet connectivity. Not at all challenging (0) vs. extremely challenging (1)		−0.269 ***		−0.027
Lack of quiet space. Not at all challenging (0) vs. challenging (1)		−0.027		−0.048 *
Lack of quiet space. Not at all challenging (0) vs. challenging (1)		−0.002		−0.130 **
Model 2	0.219 ***		0.155 ***
Medium of instruction. Kazakh (0) vs. Russian (1)		−0.316 ***		−0.136 ***
Medium of instruction. Kazakh (0) vs. mixed (1)		−0.113 ***		−0.091 ***
Not being able to see students’ reactions. Not at all challenging (0) vs. challenging (1)		−0.158 ***
Not being able to see students’ reactions. Not at all challenging (0) vs. extremely challenging (1)		−0.284 ***
Not knowing students’ learning because of collusion/plagiarism. Not at all challenging (0) vs. challenging (1)		−116 **		−0.017
Not knowing students’ learning because of collusion/plagiarism. Not at all challenging (0) vs. extremely challenging (1)		−0.332 ***		−0.148 ***
Low student motivation. Not at all challenging (0) vs. challenging (1)		−140 ***		−0.048
Low student motivation. Not at all challenging (0) vs. extremely challenging (1)		−0.244 ***		−0.103 ***
I have to explain and justify to parents the assessment of their child’s work. No (0) vs. yes (1)		−236 ***		−0.133 ***
Because I am teaching from home, parents think I am not working hard enough. No (0) vs. yes (1)		−0.250 ***		−0.251 ***
Teacher collegiality and collaboration improved during online teaching. No (0) vs. yes (1)		0.088 ***		0.067 ***
I feel under constant surveillance by the school principal/deputy director(s). No (0) vs. yes (1)				−0.286 ***
I have to constantly resolve arguments between my children over space and resources. No (0) vs. yes (1)		−0.097 ***		−251 ***
In times of crisis, we can turn to each other for support. No (0) vs. yes (1)		−0.163 ***		0.121 ***
Poor Internet connectivity. Not at all challenging (0) vs. challenging (1)		−0.184 ***
Poor Internet connectivity. Not at all challenging (0) vs. extremely challenging (1)		−0.266 ***
Lack of quiet space. Not at all challenging (0) vs. challenging (1)				−0.057 *
Lack of quiet space. Not at all challenging (0) vs. challenging (1)				−0.142 ***

Note. * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001.

), with Model 2 chosen for interpretation. The models for physical and psychological well-being met all multicollinearity assumptions. For physical well-being, the tolerance statistics ranged from 0.842 to 0.984. For psychological well-being, the tolerance statistics ranged from 0.871 to 0.983, and the VIF values for both models were below two, suggesting no multicollinearity concerns. Model 2 for physical well-being (F = 128, p < 0.001) explained 21.9% of the variation. Model 2 for psychological well-being (F = 89.5, p < 0.001) explained 15.5% of the variation.

The school medium of instruction emerged as a significant predictor of teacher well-being. Teachers in schools with a Russian medium of instruction reported worse physical (B = −0.316, p < 0.001) and psychological well-being (B = −0.136, p < 0.001) compared to those in schools with a Kazakh medium of instruction. Similarly, teachers in schools with a mixed medium of instruction reported lower physical (B = −0.113, p < 0.001) and psychological well-being (B = −0.091, p < 0.001) than their counterparts in Kazakh instruction schools. This suggests that teachers in Kazakh medium schools generally reported better well-being during the emergency remote teaching context.

Furthermore, student-related issues negatively predicted teachers’ well-being. Specifically, teachers who felt they could not see students’ reactions due to cameras being off reported poorer physical well-being. Those who found this situation ‘Challenging’ (B = −0.158, p < 0.001) or ‘Extremely Challenging’ (B = −0.284, p < 0.001) reported worse physical well-being compared to those who saw it as ‘Not a challenge at all’.

Similarly, concerns about not knowing students’ learning progress—due to parents completing tasks for children or students copying from one another—also negatively impacted both physical and psychological well-being. Teachers who viewed this issue as ‘Challenging’ (B = −0.111, p = 0.006) or ‘Extremely Challenging’ (B = −0.332, p < 0.001) reported worse physical well-being compared to those who considered it ‘Not challenging at all’. The negative impact on psychological well-being was particularly notable among teachers who considered the issue of cheating to be ‘Extremely challenging’ (B = −0.148, p < 0.001).

Another student-related negative predictor was low student motivation. Teachers who viewed this issue as ‘Challenging’ (B = −0.057, p = 0.002) or ‘Extremely Challenging’ (B = −0.084, p < 0.001) reported lower physical well-being compared to those who saw it as ‘Not challenging at all’. Additionally, teachers perceiving low student motivation as ‘Extremely challenging’ also experienced lower psychological well-being (B = −0.026, p = 0.026).

Poor relationships with parents negatively predicted teacher well-being. Teachers who needed to explain and justify assessments to parents reported lower physical (B = −0.236, p < 0.001) and psychological well-being (B = −0.133, p < 0.001) compared to those who did not feel the need to do so. Additionally, teachers who believed that parents underestimated their workload while teaching remotely also had poorer physical (B = −0.250, p < 0.001) and psychological well-being (B = −0.286, p < 0.001).

Teacher collegiality positively predicted teacher well-being. In the COVID-19 emergency, collaboration helped teachers navigate challenges and enhanced their well-being. Specifically, teachers who agreed that “teacher collegiality and collaboration improved during online teaching” reported better physical (B = 0.088, p < 0.001) and psychological well-being (B = 0.067, p < 0.001) compared to those who disagreed.

School leadership style rooted in control negatively impacted teacher well-being. Teachers who felt under continual surveillance by headteachers reported lower psychological well-being (B = −0.286, p < 0.001).

Relationships with family members were significant predictors of teacher well-being. During the emergency lockdown, families faced challenges related to sharing space and resources. Teachers who agreed that they “constantly had to resolve arguments between my children over space and resources” reported worse physical (B = −0.097, p < 0.001) and psychological well-being (B = −0.251, p < 0.001). While family support generally enhances teacher well-being, this study revealed a contradictory finding regarding physical well-being. Teachers who affirmed that “in times of crisis, we can turn to each other for support” tended to have lower physical well-being (B = −0.030, p < 0.001) compared to those who disagreed. Conversely, teachers who felt supported by their families reported better psychological well-being (B = 0.121, p < 0.001).

Resources at home and school significantly negatively predicted teachers’ physical well-being. Teachers who described “poor internet connectivity at home/school during blended learning” as ‘Challenging’ (B = −0.111, p = 0.006) or ‘Extremely Challenging’ (B = −0.332, p < 0.001) reported worse physical well-being compared to those who found the issue ‘Not challenging at all’. Additionally, teachers who perceived “a lack of quiet space at home/school during blended learning” as ‘Extremely challenging’ experienced reduced psychological well-being (B = −0.142, p = 0.006).

The findings from both the home and school levels indicate that teacher well-being is affected by various factors, including the medium of instruction, relationships with parents and their own families, teachers’ perceptions of collegiality among peers, the level of oversight from school leaders, student engagement and motivation, as well as the quality of their workspace and Internet connectivity.

6.3.3. Community-Level Predictors

Variables to measure the impact of community-level predictors on teacher well-being included responsibility and duty variables, as well as respect, recognition, and exposure to COVID-19. Model 1, which analysed both physical and psychological well-being in relation to these predictors, was significant, with all included predictors deemed relevant (

Table 10. Multiple linear regression analysis of community-level predictors and teacher well-being.

Variable	Physical		Psychological
Predictors	R2	β	R2	β
Model 1	0.128 ***		0.102 ***
Being expected to do household chores during work time. No (0) vs. yes (1)		−0.308 ***		−0.455 ***
Family obligations. Not at all challenging (0) vs. challenging (1)		−0.233 ***		−0.109 ***
Family obligations. Not at all challenging (0) vs. extremely challenging (1)		−0.494 ***		−0.434 ***
Remote work and better distribution of work and domestic responsibilities. No (0) vs. yes (1)		0.256 ***
Feeling overwhelmed by parental responsibility. No (0) vs. yes (1)		−0.400 ***
Exposure to COVID-19		−0.188 ***		−0.228 ***

Note. *** p ≤ 0.001.

). Consequently, we did not proceed with Model 2. Both the physical and psychological well-being models satisfied all multicollinearity assumptions. The tolerance statistics ranged from 0.944 to 0.988 for physical well-being and from 0.993 to 0.996 for psychological well-being, with VIF values below two for both models. For physical well-being, the model was significant (F = 191, p < 0.001) and accounted for 12.8% of the variation. Likewise, for psychological well-being, the model was significant (F = 324, p < 0.001) and accounted for 10.2% of the variation.

Multiple regression analysis revealed that teachers who agreed with the statement “My household members expect me to do household chores during school closures during my work time”, reported worse physical (B = −0.308, p < 0.001) and psychological well-being (B = −0.455, p < 0.001) compared to those who disagreed (Table 10). Additionally, teachers who viewed their family obligations as ‘Challenging’ (B = −0.233, p < 0.001) or ‘Extremely challenging’ (B = −0.494, p < 0.001) had lower physical well-being scores than those who considered them ‘Not at all challenging’. The impact of family obligations on psychological well-being was similar (B = −0.109; −0.455, p < 0.001). Furthermore, teachers who felt overwhelmed by their responsibilities as parents reported lower physical well-being (B = −0.400, p < 0.001). In contrast, those who agreed that working from home allows them to balance their work and domestic responsibilities scored higher in terms of teacher well-being (B = 0.256, p < 0.001)

Exposure to COVID-19 among teachers and their family members and friends negatively predicted both physical (B = −0.188, p < 0.001) and psychological well-being (B = −0.228, p < 0.001).

6.3.4. National-Level Predictors

National-level predictors of teacher well-being during emergency remote teaching significantly predicted both physical and psychological well-being (

Table 11. Multiple linear regression analysis of national-level predictors and teacher well-being.

Variable	Physical		Psychological
Predictors	R2	β	R2	β
Model 1	0.0067		0.014
Freedom to choose online platforms. No (0) vs. yes (1)		0.122 ***		0.166 ***
Opportunities to network. No (0) vs. yes (1)		0.061 ***		0.087 ***
Freedom to revise curriculum. No (0) vs. yes (1)		0.090 ***		0.079 ***
Top-down approach in decision making. No (0) vs. yes (1)		−0.102 **		−0.167 ***

Note.** p ≤ 0.01; *** p ≤ 0.001.

). Both the physical and psychological well-being models satisfied all multicollinearity assumptions. The tolerance statistics ranged from 0.822 to 0.990 for physical well-being and from 0.822 to 0.990 for psychological well-being, with VIF values below two for both well-being models. Model 1, which examined physical well-being in relation to national predictors, was significant (F = 21.8, p < 0.001) and accounted for 0.67% of the variation (Table 11). All included predictors were found to be significant. Model 1 for psychological well-being in relation to national predictors was also significant (F = 83.589, p < 0.001) and accounted for 1.4% of the variation (Table 11).

Teacher autonomy and networking opportunities positively impacted their well-being. Specifically, the freedom to choose online platforms was linked to improved physical well-being (B = 0.122, p < 0.001) and psychological well-being (B = 0.166, p < 0.001). Additionally, the ability to revise the curriculum positively predicted both physical (B = 0.090, p < 0.001) and psychological well-being (B = 0.079, p < 0.001). Similarly, networking opportunities were also associated with enhanced physical (B = 0.061) and psychological well-being (B = 0.087). Conversely, teachers who reported a lack of autonomy and involvement in decision making and who agreed with the restrictions imposed by top-down management experienced worse physical (B = −0.102, p < 0.001) and psychological well-being (B = −0.167, p < 0.001) compared to those who did not experience top-down management.

Both well-being models are statistically significant but explain very little variance (R² values below 2%), suggesting the predictors have limited explanatory power for well-being outcomes. The strongest positive predictor for both physical and psychological well-being is “Freedom to choose online platforms”. “Top-down decision-making” has a consistent and negative effect on both well-being measures, particularly psychological well-being.

7. Discussion

Overall, Kazakhstani teachers reported poor physical well-being, primarily due to eyesight issues, aligning with studies that show a decline in teachers’ health during school closures [20,21,32,54]. In contrast to previous research [20,29,35,50], they expressed a more positive view of their psychological well-being. This reluctance to acknowledge issues like depression may stem from the direct nature of the questions. Interestingly, similar to our findings, Herman et al. noted that teachers in the US reported lower stress levels at the pandemic’s onset, likely due to the end of the school year [57]. In Kazakhstan, data collection began in May 2021, coinciding with the academic year’s end and summer holidays, which may have contributed to reduced stress. However, many teachers chose “not applicable” or “disagree” regarding their psychological well-being, suggesting they either do not see mental health issues as relevant to themselves or are hesitant to reflect on them, possibly due to social desirability bias [78]. It is also important to note that Kazakhstani teachers did not face certain risk factors, such as salary deductions, which occurred in some contexts [50,91]. On the contrary, the government significantly increased teacher salaries during school closures to boost morale [58]. Additionally, after facing challenges in the transition period, the government later implemented a comprehensive strategy that included professional development for teachers, adjustments to school timetables, and revised assessment policies aimed at enhancing teacher well-being. These proactive measures may have further contributed to safeguarding the well-being of teachers during this challenging period.

While we lack a direct comparison base, as no other studies have examined such a comprehensive range of protective and risk factors associated with teacher well-being within a single study, the regression models at each of the four levels of the socio-ecological framework were statistically significant. Specifically, at the individual level, the predictors—some of which are shown in the green circle in Figure 1—accounted for 9.3% and 4.5% of the variance in physical and psychological well-being, respectively. At level 2, school and home, the predictors—illustrated in the orange circle in Figure 1—accounted for 21.4% and 15.5% of the variance, respectively. At level 3, community, the predictors—depicted in the blue circle in Figure 1—explained 12.8% and 10.2% of the variance in physical and psychological well-being, respectively. Finally, at level 4, the national level, the predictors—represented in the pink circle in Figure 1—accounted for 0.6% and 1.4% of the variance in physical and psychological well-being, respectively.

Similar to Ainsworth and Oldfield’s study [51], our findings indicate that factors related to school, home, and community are more influential than individual characteristics when predicting teacher well-being. This highlights the significance of environmental influences on teachers’ well-being, which is a crucial and policy-relevant finding of our study. This finding aligns with critiques of “hyper-individualisation” in teacher resilience discourse, which risks shifting responsibility onto teachers without addressing systemic challenges [92], p. 707. While the COVID-19 pandemic is no longer an emergency, the persistent pressures of globalisation and neoliberalism remain significant risk factors for teachers, raising concerns about the impact of resilience rhetoric on their identities and coping abilities [1,2]. Our study echoes the conclusions of studies on teacher well-being that advocate for greater attention to contextual and organisational factors in supporting teacher well-being beyond merely focusing on individual characteristics [26,43,77].

7.1. Well-Being and Individual Characteristics

Individual-level predictors significantly influenced teachers’ physical and psychological well-being, with self-efficacy in online teaching positively impacting both. Teachers with higher confidence in their online abilities reported better physical and psychological outcomes, supporting previous research on the positive effects of self-efficacy [37,40,57].

Negative predictors of teacher well-being included gender, age, and educational qualifications. Female teachers reported poorer physical and psychological well-being than males, aligning with prior research on the health and mental well-being of female teachers during the COVID-19 pandemic [20,27,35,37,40,48,50,52,53,54], including findings specific to Kazakhstan [70].

Older teachers reported worse physical well-being, likely due to negative effects from excessive screen time and prolonged sitting, consistent with existing research [53]. However, the negative relationship between age and psychological well-being contradicts some research suggesting that older, more experienced teachers are less anxious [40]. In the context of the COVID-19 pandemic, older teachers may experience increased anxiety about online teaching. Additionally, studies have shown that younger teachers generally have better psychological well-being than their older counterparts [18,27,31,35,39,40,52].

Surprisingly, higher levels of education negatively predicted teachers’ physical and psychological well-being, with those holding Bachelor’s and postgraduate degrees reporting worse physical health than their peers with lower or vocational education. This may be due to increased responsibilities that lead to higher workloads and stress, or it could stem from the greater expectations that highly qualified teachers have regarding their job satisfaction [93].

Ethnicity influenced the well-being of Kazakhstani teachers, with non-Kazakh teachers reporting poorer physical and psychological outcomes. Although specific studies on teachers’ well-being by ethnicity are limited, some research suggests that Black teachers may have better psychological well-being than White teachers due to more effective coping strategies [29,53]. Additionally, the self-reported better physical well-being of Kazakh teachers may be influenced by social desirability bias, indicating a reluctance to openly criticise their situations [78]. However, we argue that culturally, Kazakh people are influenced by the concept of “tau ba”, or “shukur”, which signifies satisfaction or contentment with one’s current situation and is associated with positive well-being [94]. This cultural perspective may contribute to the positive subjective well-being reported by Kazakh teachers.

Interestingly, number of children positively predicted teachers’ self-reported well-being, as having more children was linked to improved physical and psychological health. This may be due to the sense of purpose and fulfilment children bring, enhancing emotional resilience and life satisfaction. In Kazakhstan, motherhood is viewed as a key aspect of ideal womanhood [95], and the government promotes childbirth through various pronatalist policies [96]. Additionally, engaging in physical activities with children can improve physical health, while parenting support networks—such as shared responsibilities and emotional backing—can reduce stress and enhance coping mechanisms.

While income was not related to psychological well-being, surprisingly, teachers with higher incomes reported significantly lower levels of physical well-being compared to their counterparts with below-average incomes. This trend may be attributed to the increased job-related stress or higher work demands faced by those in higher income brackets. Additionally, it is important to note that individuals with higher incomes are often more experienced and, therefore, older, which could also contribute to their reported lower levels of physical well-being.

7.2. Well-Being and School and Home Characteristics

School- and home-level factors significantly impacted the physical and psychological well-being of Kazakhstani teachers. The medium of instruction was a notable negative predictor; teachers in Russian and mixed-medium schools reported worse well-being than those in Kazakh-medium schools. This indicates that teachers in Kazakh-medium schools generally fared better during remote teaching, possibly due to a higher proportion of ethnic Kazakh teachers who reported better well-being than their non-Kazakh counterparts. Additionally, social desirability bias may play a role, as Russian-speaking teachers might be more inclined to express criticism.

Student-related issues also impacted Kazakhstani teachers’ well-being. Teachers who couldn’t see students’ reactions due to turned-off cameras reported poorer physical health, emphasising the importance of student engagement in the learning process [41,59]. Concerns about students’ learning progress and low student motivation also contributed to decreased well-being, aligning with international evidence [12,46,59,60]. This indicates that fostering student engagement and motivation is essential not only for student learning but also for supporting teachers’ well-being.

Relationships with parents also impacted teacher well-being similar to existing research [47]. Teachers who needed to explain and justify assessments to parents reported lower physical and psychological well-being than those who did not. Additionally, those who felt parents underestimated their remote teaching workload also experienced poorer well-being. These findings highlight the importance of effective communication and fostering positive relationships with parents, which can significantly enhance teacher well-being.

Collegiality positively predicted teacher well-being, aligning with existing studies [12,57]. Collaboration enabled teachers to navigate challenges, with those who felt collegiality improved during online teaching reporting better physical and psychological health than those who disagreed. This suggests that the sense of community and shared purpose fostered by collegiality significantly enhances teachers’ resilience and overall well-being during difficult times.

Likewise, supporting existing studies [59,63], a directive school leadership style negatively predicted teacher well-being, with teachers who felt constantly monitored by headteachers reporting lower physical and psychological well-being. Teachers thrive in environments where they feel trusted and empowered, indicating that prioritising trust and support in leadership can significantly enhance their well-being.

Family relationships significantly influenced teacher well-being during the emergency lockdown. Teachers who frequently resolved conflicts over space and resources reported poorer physical and psychological health. While family support typically enhances well-being [26,52], this study found that those who felt they could rely on family during crises reported lower physical well-being, although they experienced better psychological well-being. This discrepancy may stem from the time and energy required to seek support, highlighting the complexities of family dynamics and their effects on teacher well-being.

Resources at home and school significantly predicted teacher well-being, with poor Internet connectivity and the absence of a quiet workspace leading to worse physical and psychological health. Existing studies also indicate that inadequate digital infrastructure and shared spaces increase teachers’ stress and anxiety, underscoring the need for adequate resources and environments to support their effectiveness and well-being [42,48].

7.3. Well-Being and Community Characteristics

The blurring of boundaries between home and work adversely affected teachers’ physical and psychological well-being, aligning with existing studies [26,50,66,97]. Teachers who were expected to handle household chores during work hours and those who found family obligations “challenging” or “extremely challenging” reported poorer well-being, with feelings of being overwhelmed by parental responsibilities linked to lower health, supporting existing findings [56]. In contrast, teachers who felt remote teaching helped them balance work and domestic duties reported improved well-being.

While many studies indicate that female teachers are at greater risk for adverse well-being, few explore the reasons behind this, as noted by Ma et al. [40] in their systematic review of 54 quantitative studies on teacher well-being during the pandemic. The current study addresses this gap by identifying additional predictors, such as domestic responsibilities, childcare, and parenting, which disproportionately burden women and contribute to their increased risk of deteriorating physical and psychological well-being. Collectively, these findings underscore the significant burden that the unequal distribution of domestic chores, childcare, and caregiving responsibilities places on women in paid employment, including teachers [50,79].

Additionally, consistent with previous research [48,67], exposure to COVID-19 among teachers, their family members, and close community members was identified as a negative predictor of well-being.

7.4. Well-Being and National Characteristics

National-level contextual variables also influenced teachers’ physical and psychological well-being, but the models’ low explanatory power suggests other significant factors are also at play. Teacher autonomy and networking opportunities were key contributors to well-being, with autonomy in selecting online platforms and revising the curriculum linked to improved health. These findings align with previous research indicating that autonomy-supportive leadership and collegial collaboration enhance psychological well-being [19,28,30,62,63,67]. However, only 17.6% of Kazakhstani teachers felt they had this autonomy, highlighting the contradiction between teachers’ experiences and school leaders’ perceptions of their autonomy practices [80]. Conversely, teachers experiencing a lack of autonomy and top-down management reported poorer well-being, a prevalent issue in Kazakhstani schools [72]. Furthermore, professional development in online teaching did not correlate with teacher well-being. This aligns with the findings of Cann et al., which indicated that although teacher professional development interventions enhanced teacher efficacy, they did not significantly reduce burnout and were generally ineffective in improving teachers’ psychological well-being [26].

7.5. Limitations

The findings of this study should be interpreted in light of its limitations. First, the measurement scale used was specifically developed for the context of remote schooling during the COVID-19 pandemic and has not been validated. Second, since data were collected solely in Kazakhstan, which has a unique post-Soviet background, the findings may not be easily comparable to those from countries with vastly different cultural contexts. Third, well-being scores were self-reported by the participating teachers rather than being objectively measured. Fourth, although the sample size is substantial, the use of a non-probability sampling method limits its representativeness and, consequently, the generalisability of the results. Finally, the cross-sectional design of this study restricts the ability to draw causal inferences.

7.6. Implications

7.6.1. Multi-Foci and Systemic Interventions

While individual resilience is important for enhancing teacher well-being, it should not be viewed as the only solution to the stresses teachers encounter. Systemic changes are essential to create supportive environments that promote well-being. Policymakers and educational leaders must prioritise comprehensive strategies that address both individual and contextual factors. Multi-faceted interventions, such as mindfulness training, positive psychology [11], professional development opportunities, fair compensation [24], supportive management, manageable workloads, networking opportunities, and positive school cultures that foster collaboration and strong organisational relationships [44,51,77] have been shown to effectively reduce stress and burnout while promoting job satisfaction and overall well-being, compared to single-focus interventions [26].

Addressing the complex challenges of teacher well-being requires a collaborative and strategic approach that incorporates individual, relational, and contextual factors rather than relying on short-term fixes [77]. A cultural shift prioritising teacher well-being across various systems and stakeholders is essential to ensure that initiatives are sustainable and promote long-term flourishing for teachers. By embedding well-being into teacher preparation, induction, mentoring, and professional learning, we can create a more sustainable framework that supports teachers in their roles and acknowledges the broader social and environmental pressures impacting their professional lives [77]. Ultimately, fostering a holistic understanding of teacher well-being will contribute to a healthier educational ecosystem that benefits teachers, students, and communities alike.

7.6.2. Autonomy-Supportive Leadership

Regardless of the presence and nature of the emergency, schools can enhance their effectiveness and support teacher well-being by involving teachers in decision making and granting them greater autonomy without any extra funding. However, in contexts like Kazakhstan, achieving this may require a significant cultural shift away from entrenched top-down leadership rooted in Soviet legacies.

7.6.3. Social Support

Promoting mentoring programmes and peer support networks, both face-to-face and virtual, along with providing time and flexibility for collaboration, can enhance teachers’ perceived social support and strengthen their resilience during challenging times. Targeted school-based support can significantly enhance teacher agency, recognising their expertise, allowing experimentation, and fostering collegial support [98].

7.6.4. Strengthening Social–Emotional Resources

To enhance teacher well-being during both crises and normal circumstances, schools should implement targeted professional development programmes that focus on building stress management, emotional regulation, and resilience training, while also promoting a supportive work environment. Additionally, providing access to mental health resources and creating opportunities for self-care can further empower teachers to maintain their well-being.

7.6.5. Equitable Resources

To enhance teacher well-being in both crises and normal circumstances, equitable access to essential teaching resources—such as reliable Internet, quality learning materials, suitable platforms, and adequate workspace—is crucial. Providing these tools empowers teachers to perform effectively and ensures a supportive environment, ultimately benefiting both educators and students.

7.6.6. Transforming Gender Norms for Equity

Transforming inequitable cultural and gender norms is essential for supporting female teachers and women in the workforce. By implementing policies that promote shared household responsibilities, such as paid parental leave for both parents and accessible childcare services, alongside public awareness campaigns that challenge traditional gender roles and flexible work arrangements, the burdens of domestic chores and caregiving on women can be alleviated in the long term.

7.6.7. Professional Development in Digital Literacy and Pedagogy

While this study did not find professional development (PD) in online teaching to be a protective factor of teachers’ psychological, effective PD can enhance teachers’ self-efficacy and teaching practices and foster positive student outcomes [99]. In light of increasing challenges from environmental crises, infectious disease outbreaks, and advancements in artificial intelligence, it is essential to provide teachers with effective PD that enables them to effectively utilise digital and ICT tools in their teaching.

7.6.8. Future Research

Our findings reveal both methodological and substantive gaps. Methodologically, the limitations of a cross-sectional design restrict our ability to infer causation and directionality [51]; therefore, future studies should consider using longitudinal designs to produce more robust findings. Additionally, research in authoritarian contexts raises safety concerns for both researchers and participants [100], often leading to silenced voices and a reluctance to share experiences openly. In such politically closed environments, it is essential to broaden methodological approaches beyond surveys and interviews by incorporating creative, visual, and multimodal methods, such as mapping, sorting and ranking exercises, photovoice, community-based participatory video, and participatory theatre, to mitigate social desirability pressures and foster an enabling environment for research participation [101].

In substantive terms, this study is one of the few that investigates the association between the medium of instruction and teacher well-being. While existing research highlights disparities in student test scores among different medium of instruction schools in both Kazakhstan [72,102] and other contexts [16,103], our findings suggest that these differences may also extend to organisational structure and school culture. Therefore, further exploration of these areas is essential for developing strategies to promote equity across various mediums of instruction in schools. Additionally, while improving relationships with colleagues and students is positively associated with teacher well-being, Cann et al. note that very few studies have focused on examining such interventions [26]. Therefore, future research should explore the impact of interventions aimed at enhancing organisational relationships on measures of teacher well-being.

8. Conclusions

This study advances the field by empirically testing Falk et al.’s [24] conceptual model of teacher well-being in emergencies, which had not been previously investigated. It expands the framework to include the COVID-19 school closures, utilising a substantial sample of over 1900 teachers to demonstrate how a comprehensive model can reveal the diverse factors influencing teacher well-being across four socio-ecological levels: individual, school and home, community, and national. The findings highlight the importance of contextual factors, in addition to individual teacher characteristics, in affecting teacher well-being, as well as the complex interplay between cultural and contextual influences and subjective perceptions of well-being. As a rare study integrating such an expansive and comprehensive framework, the findings underscore the critical importance of understanding the social, political, and organisational contexts that shape teacher well-being. This understanding can inform policymaking and practices aimed at developing sustainable and culturally appropriate solutions to protect teacher well-being.

While the findings are specific to the COVID-19 context and Kazakhstan’s unique social and educational landscape, many significant predictors are likely relevant in non-emergency environments and other contexts. For instance, even after schools have reopened, factors such as teacher autonomy, leadership style, access to adequate resources, manageable workloads, equitable distribution of domestic responsibilities, self-efficacy, and student motivation are expected to continue influencing teacher well-being in the post-COVID-19 landscape. By addressing the identified risk factors and enhancing the protective factors, policymakers and school leaders can effectively support and improve teacher well-being. This proactive approach not only benefits teachers but also contributes to a more positive educational environment, ultimately leading to better outcomes for students.