1 Introduction

Over the past two decades, there has been a growing focus on researching teacher wellbeing, resulting in numerous empirical studies from diverse disciplines (Hascher & Waber, 2021). However, significant gaps still exist in our understanding of teacher wellbeing. Teacher wellbeing is often ill-defined, if at all (Cann et al., 2023b; McCallum et al., 2017). In this study, we adopt the term ‘educator wellbeing’, which encompasses the wellbeing of teachers as well as other staff members within a school, recognizing their collective importance in fostering positive school environments (McCallum, 2021). We define educator wellbeing as educator satisfaction with their work, a sense of engagement in a profession that contributes to others, supportive working relationships, and low levels of stress (Cann et al., 2022).

Our definition of educator wellbeing positions the concept not only as an individual state, but also a relational and organisational construct, encompassing relationships at work and engagement in the school community. However, interventions to improve educator wellbeing often neglect the relational and organisational dimensions, and instead focus on the individual (Cann et al., 2023b; Waters & Loton, 2019). The field of positive psychology, the science of wellbeing, has faced criticism for its focus on the individual, both from within and outside the field (van Zyl et al., 2023). However, there are changes towards a greater consideration of the relational and contextual influences on wellbeing as the field evolves into its ‘third wave’ (Lomas et al., 2021) and develops ‘systems informed positive psychology’ (Kern et al., 2020). Other conceptualisations of wellbeing, such as Bronfenbrenner’s ecological theory (1979) already consider factors beyond the individual and have been used to explore the wellbeing of pre-service teachers (Price & McCallum, 2015). However, the ecological approach is seldom used in developing educator wellbeing interventions, and positive psychology informed school-based wellbeing programs have been criticised for overlooking contextual influences on wellbeing (Ciarrochi et al., 2016). There are also concerns that many school wellbeing programs are not scrutinised at all. A review of over 200 school-based wellbeing programs showed that over half had low-quality evidence to support their use as no studies had been conducted to evaluate their effectiveness (Dix et al., 2020). There is a need for more studies of school-based wellbeing programs, particularly those that explicitly address the relational and contextual influences on wellbeing.

We argue that adaptation is a crucial element of programs that successfully account for the contextual influences on educator wellbeing. Adaptation involves modifying programs to align effectively with the implementation context and meet the needs of recipients and communities. This requires an in-depth exploration and understanding of the context in which it is to be implemented in order to adapt the program accordingly. Thus, we position our research in the emergent third wave of positive psychology which emphasises the importance of context (Lomas et al., 2021). We examine educators’ socio-ecological context to design and adapt an intervention, utilizing qualitative methodologies to present a case study of adapting the intervention to a specific school. The intervention is an educator wellbeing program that attends to individual, relational, and contextual influences on wellbeing—the MARKERS (Multiple Action Responsive Kit for Educator, Relational, and School wellbeing) Program. The case study illustrates how we examined and adapted to the context in one school in Aotearoa New Zealand, and the outcomes achieved by the educator wellbeing program.

2 Literature Review

The seminal paper on positive psychology describes it as a science that seeks to “understand and build the factors that allow individuals, communities, and societies to flourish.” (Seligman & Csikszentmihalyi, 2000, p.5). However, despite this definition advocating for an understanding of factors contributing to community flourishing, positive psychology has predominantly focused on the individual (van Zyl et al., 2023). Extensive evidence supports the relationship between wellbeing, or reduced ill-being, and individual traits such as optimism (Alarcon et al., 2013) and enthusiasm (Sun et al., 2018), as well as habits like physical activity (White et al., 2017) and meditation (Goyal et al., 2014). These findings inform numerous positive psychology interventions that target individuals. For instance, interventions may be tailored to specific populations based on individual characteristics such as depression levels or centred around individual actions like practising gratitude or mindfulness. In recent years, the positive psychology field has witnessed a shift towards incorporating more relational and contextual approaches (Kern et al., 2020; Lomas et al., 2021). In the following sections, we examine relational and contextual influences on wellbeing, highlight the necessity of integrating them into wellbeing interventions, and discuss the approaches available to contextualise and adapt interventions in schools.

2.1 Relational and Contextual Influences on Wellbeing

An individuals’ relationships and context—their socio-ecological environment—influences their wellbeing (Reupert, 2017). Social interactions, for example, are linked to greater wellbeing, happiness, and positive affect (Sandstrom & Dunn, 2014; Sun et al., 2020; Watson et al., 1992). Conditions within a community or society have also been shown to influence wellbeing. The level of trust within a society is linked to subjective wellbeing (Helliwell & Wang, 2010), and organisational drivers of wellbeing at work include trust in leadership, and employee involvement, recognition, and reward (Lomas, 2019). Increasingly these relational and contextual influences are being recognised in the field of positive psychology, as described in ‘third wave’ positive psychology (Lomas et al., 2021) and ‘systems informed positive psychology’ (SIPP: Kern et al., 2020). SIPP calls for positive psychology to draw on concepts from systems science, including more use of qualitative methodologies, to move positive psychology “from a science of individual wellbeing to a science of collective benefit” (Kern et al., 2020, p. 714). Third wave positive psychology describes the start of a shift towards “going beyond the individual person as the primary focus of enquiry to look more deeply at the groups and systems in which people are embedded” (Lomas et al., 2021, p. 660).

Despite this increasing awareness of the importance of relational influences on wellbeing, these ideas do not feature prominently yet in positive psychology research. For example, Brauer and Proyer (2023) highlight the lack of research exploring the link between romantic relationships and wellbeing—only 3.7% of studies published in leading positive psychology journals focused on this topic, and in particular there are a paucity of studies using dyadic data. Similarly, there are few studies exploring workplace relationships and their impact on wellbeing. For example, although social network analysis is often used to explore relational ties within workplace networks (e.g. Cross & Parker, 2004), enabling detailed analyses of dyads and their impacts, few workplace social network studies consider the links to wellbeing. Exceptions to this include social network studies of burnout contagion in teachers (Kaihoi et al., 2022; Kim et al., 2017) and wellbeing contagion in teachers (Cann et al., 2022).

The importance of relational and contextual influences on wellbeing has also yet to consistently underpin interventions to improve wellbeing. Positive psychology interventions (PPIs) are often grounded in theories that include a relational aspect, as evident in many wellbeing frameworks that illustrate the elements of wellbeing. For example, relationships are part of the PERMA framework (Positive Emotion, Engagement, Relationships, Meaning, and Accomplishment: Seligman, 2011), and positive relationships are one of Ryff’s six dimensions of wellbeing (Ryff, 1989) and one of the 10 features of flourishing (Huppert & So, 2013). However, although some interventions focus on improving relationships (for example, active construct responding: Shankland & Rosset, 2017), they are primarily viewed in terms of their benefits for individuals (Kern et al., 2020). It is not uncommon for PPIs to be conducted within organisations such as schools and workplaces—for example, Carr et al.’s (2021) meta-analysis puts the figure at 44%. However, it is rare that PPIs in organisations target group-level outcomes, with most targeting individuals (van Woerkom, 2021). Furthermore, interventions often span multiple organisations rather than being specific to one, and when interventions do occur within a single organisation, studies rarely detail how the intervention was adapted to the unique organisational context (Cann et al., 2023b). For example, Donaldson et al. (2021a) identified adaptability as a key criterion for exemplary PPIs but primarily focused on adapting interventions to individuals (e.g. participants choose content or when they access content) or general socio-economic contexts (e.g. offering interventions via phone or mail for individuals lacking technology or internet access). Even within the area of Positive Organisational Scholarship (POS), described as the organisational equivalent of positive psychology (Caza & Cameron, 2008), the degree to which interventions are targeted to a particular organisational context is variable. For instance, there are exemplar case studies illustrating the use of POS in designing interventions tailored to specific higher education organisations (Cameron, 2021). In contrast, Donaldson et al. (2021b) description of Positive Organisational Psychology Interventions (POPIs) include studies such a gratitude intervention with teachers from multiple schools, which lacked adaptation to their specific circumstances or the education field as a whole (Chan, 2010). Throughout this study we argue that interventions that target the organisational level should include adaptations to the particular organisational context in which they are implemented.

2.2 Wellbeing Interventions in Schools—Understanding and Adapting to Contextual Factors

Positive education is defined as “the application of positive psychology in educational contexts to promote positive mental health, allowing students, staff and the wider school community to flourish” (Hoare et al., 2017, p. 57). However, the wellbeing of school staff is often neglected in positive education interventions, with only 2% of interventions taking a whole-school approach that includes building staff wellbeing (Waters & Loton, 2019). Positive education is often seen as an isolated program, lacking integration into school strategies and the broader educational context (White, 2017), and prioritising content whilst paying insufficient attention to context (Ciarrochi et al., 2016; Halliday et al., 2019). Part of this issue may be due to the dominance of experimental methods, particularly randomised controlled trials (RCTs), in positive education research (Waters & Loton, 2019) and psychology as a whole (Diener et al., 2022). Experiments rarely provide rich situational descriptions, limiting our understanding of how contextual factors influence outcomes (Diener et al., 2022). Studies of interventions to improve educator wellbeing seldom describe the particular school context in which they are conducted, often due to recruiting educators from numerous schools. Even where interventions take place within a single school, descriptions of context are extremely brief (Cann et al., 2023b). Research methodologies such as case studies can offer valuable insights into contextual influences by providing in-depth descriptions and analysis of the context in which interventions are conducted. However, there are few case studies of best practice in positive education, especially in disadvantaged and indigenous cultures (White, 2017). Therefore, more case study research is essential in order to understand the effectiveness of positive psychology interventions in real-world settings.

Case studies provide an opportunity for in depth explanations of how interventions have been adapted to a particular context. Occasionally psychological interventions may adapt content to be more suitable for educators, such as incorporating modules on classroom management (Cann et al., 2023b), but few studies describe adaptations to a particular school setting (for exceptions see: Halliday et al., 2019; Hoare et al., 2017). The school context, including factors like trust and collaboration, has been shown to have a significant association with educator wellbeing (Cann et al., 2022). Therefore, understanding the specific context of a school can influence the success of an intervention at that school. For example, if levels of trust are low the intervention’s success at increasing wellbeing may be limited unless it increases trust.

2.3 Lessons from Implementation Science

Implementation science is an area focussed on understanding and explaining what makes interventions work in real-world contexts (Kelly, 2012). The real-world effectiveness of a program is not solely determined by the efficacy of evidence-based practices it employs, but also the way it is implemented, especially its adaptation to a particular context. As Lipsey (2009) explains “a well-implemented intervention of an inherently less efficacious type can outperform a more efficacious one that is poorly implemented” (p. 127). To avoid poor implementation researchers often focus on fidelity, the adherence to a program curriculum (Pettigrew et al., 2015). Yet, fidelity frequently coexists with adaptation—the modification of programs to meet the needs of recipients and communities. Both fidelity and adaptation are associated with improved program outcomes (Durlak & DuPre, 2008). Therefore, researchers should strive to strike the right balance between fidelity and adaptation to maximize the impact of interventions in real-world settings.

Implementation science can provide insights into exploring how and why interventions do, or do not, work in different school contexts (Halliday et al., 2019). Two case studies illustrate how implementation science principles have informed positive education interventions (Halliday et al., 2019; Hoare et al., 2017). Halliday et al. (2019) identify contextual factors that influenced the implementation of a positive education program in one school, including: recipient, provider, organisation, intervention, and socio-cultural and political contexts. For example, organisational factors encompass social relationships, norms, and the alignment of an intervention with organisational goals (Halliday et al., 2019). Hoare et al. (2017) describe how implementation science informed their approach to embedding positive education in one particular school context. Both case studies describe the balance between fidelity and adaptation to the school context. However, such case studies are rare when it comes to wellbeing programs in schools, and most programs remain unstudied (Dix et al., 2020). There is a need for more research into wellbeing programs in schools, not only to evaluate their effectiveness, but also to illustrate effective adaptation to real-world contexts.

While this study is not an implementation study, it draws upon insights from the implementation science literature to inform the design and implementation of the MARKERS program. In particular, we considered key factors that influence intervention success, such as adaptability and suitability for the context.

2.4 Aim of this Study

This study aims to examine the design and adaptation of an intervention tailored to the specific context of a single school. We provide a detailed description of the design of an intervention to improve educator wellbeing, and present a case study of its implementation. The design includes the development of a detailed theory of change for the intervention. The case study illustrates the process of adapting the intervention to one school context and the outcomes that were achieved. Finally, we provide takeaways for researchers, practitioners, schools, and educators about school-based educator wellbeing programs.

3 Design Process—Developing a Theory of Change

To design the MARKERS program we developed a theory of change—an “approach for making underlying assumptions explicit, and using the desired outcomes of a project as a mechanism to guide planning, implementation, and evaluation” (Reinholz & Andrews, 2020, p. 3). Table 1 provides an outline of the components of a theory of change, and we will elaborate on each area as it relates to the design of the MARKERS program.

Table 1 Components of a theory of change
Full size table

3.1 A Description of the Context

The program was developed for use in schools in Aotearoa New Zealand, therefore we outline relevant aspects of the country and its education system. Aotearoa New Zealand, located in the South Pacific, has a population of approximately 5 million people from many ethnic origins: European (70.2%), Māori (16.5%), Pacific peoples (8.1%), Asian (15.1%), and others (1.7%) (Stats, 2020). Although the population is multicultural, Aotearoa New Zealand is often described as a bicultural nation, as it is founded on Te Tiriti o Waitangi, a treaty signed in 1840 seeking to form a partnership between Māori (the indigenous people) and the Crown (the British colonists).

Schools are expected to embrace Te Tiriti o Waitangi’s principle of partnership as explicitly stated in the current New Zealand Curriculum (Ministry of Education, 2007). Furthermore, there is an emphasis to engage with mātauranga Māori (Māori knowledge) as a curriculum refresh initiated in 2021 places it at the centre of all learning (Ministry of Education, 2022b). In the case of wellbeing in the curriculum, the Māori concept of hauora (wellbeing), and the Te Whare Tapa Whā model of hauora (Durie, 1985), have long been an explicit part of the health and physical education curriculum (Ministry of Education, 1999). The model is visualised as a house with four walls, each representing a dimension of wellbeing: taha tinana (physical wellbeing), taha hinengaro (mental and emotional wellbeing), taha whānau (social wellbeing), and taha wairua (spiritual wellbeing). Each wall is needed to hold up the roof, providing strength and symmetry, and capturing a holistic view of wellbeing. Educators in Aotearoa New Zealand are generally familiar with mātauranga Māori and the concept of hauora (wellbeing) through their engagement with the curriculum. However, the extent to which Māori tikanga (customs and practices) are embraced in schools can vary considerably. Given the bicultural context, any wellbeing program offered to schools should therefore also honour Te Tiriti and incorporate mātauranga Māori and tikanga as appropriate for each school.

The notion of adapting to context is widely embraced within the education system in Aotearoa New Zealand. The New Zealand Curriculum serves as a framework for designing a localized curriculum, encouraging schools to create a unique curriculum that is responsive to the needs of their community (Ministry of Education, 2019). Teachers are advised that rather than adhering to a generic set of standards or teaching practices they should develop ‘adaptive expertise’ where they modify their practice in response to their particular classroom context (Timperley et al., 2017). This approach to adaptation extends to schools designing their own local wellbeing curriculum (Ministry of Education, 2022a). However, while student wellbeing is prioritised in policy documents, teacher wellbeing is neglected (Higgin & Goodall, 2021). Considering these perspectives, it is important that a wellbeing program for educators is adaptable to context.

The MARKERS program design and implementation coincided with the COVID pandemic. During 2020 and 2021 nationwide lockdowns, schools pivoted to provide online learning to students. School sites were only open to ‘essential workers’ children, and those that could not participate in online learning at home. Continuing COVID restrictions in early 2022 lead to schools limiting visitors, and many operated ‘bubbles’ which limited the number of staff and students that could be in physical proximity. As the program design commenced during the initial stages of the COVID pandemic it was important to include options for delivery either online or face to face depending on the COVID restrictions.

The disruptions caused by the COVID pandemic added even further to educators’ workloads. Principals described feeling “stretched and stressed” (NZEI Te Riu Roa, 2023a, para. 2), and unions campaigned for understaffing to be addressed and teachers to be valued (NZEI Te Riu Roa, 2023b). There has been long-standing dissatisfaction with working conditions and the state of education, as evidenced when 50,000 kindergarten, primary and secondary school teachers and principals went on strike in March 2023 (Stuff, 2023). In light of the stress described by many in the education workforce it is important for an educator wellbeing program to include stress reduction techniques.

3.2 Overarching Program Rationale

The importance of relational and contextual influences on educator wellbeing drove the development of the program, and its adaptation to context. This led us to envision a multi-level and mutli-foci program.

Multi-level refers to activities aimed at the individual, relational, and organisational (school) level. These levels align to Bronfenbrenner’s ecological theory, encompassing the individual and their microsystem (environment), mesosystem (interrelationships) and exosystem (organisational). The ecological theory is useful for thinking about the interactions between and within levels, for example between educators’ productive working relationships and an educator’s self-efficacy (Price & McCallum, 2015). By targeting multiple levels the program aims to impact a variety of influences on wellbeing.

Multi-foci refers to the inclusion of a variety of activities, as opposed to focusing solely on one aspect, such as a mindfulness intervention. This approach allows us to address individual, relational, and organizational levels while providing adaptability through activity choices. For example gratitude practices primarily target the individual, whereas active constructive responding is relationship focused. Research demonstrates that combining multiple positive psychology interventions (PPIs) within a single intervention leads to greater improvements in wellbeing compared to using a single PPI (van Agteren et al., 2021; Donaldson et al., 2021a). Our systematic review of interventions to improve educator wellbeing also suggested that multi-foci interventions hold promise for improving educator wellbeing (Cann et al., 2023b).

3.3 Activities, Outcomes and Indicators

We selected activities for the program using a process of backwards mapping from the intended outcomes, a typical approach in theory of change design (Reinholz & Andrews, 2020). We began with the long-term outcome of the program, improved educator wellbeing, and worked backwards to determine the intermediate outcomes (achieved over the course of the program), the immediate outcomes (for each program session), and the activities used in the program. The 4I framework of organisational learning (Crossan et al., 1999) guided our selection of activities to achieve the outcomes. This framework outlines four processes—intuiting, interpreting, integrating, and institutionalizing—which align to our focus on the individual, relational, and organizational levels (refer to Table 2 for details). To plan the activities we drew on our experience as educators in designing curriculum, grounded in adult learning theory (Merriam, 2008). Additionally, the activities also align to the five components of exemplar multi-component positive psychology interventions described by Donaldson et al. (2021a): learn, practice, reflect, relate, and plan.

Table 2 The 4I Framework
Full size table

The theory of change diagram in Fig. 1 captures the activities and outcomes of the program. It presents a generic description for each activity and outcome, rather than providing content specific examples for each session. The arrows in the diagram represent links between the activities and the immediate, intermediate, and long-term outcomes. For example, the activities ‘encounter information and ideas’ and ‘discussion’ link to the immediate outcome of building wellbeing knowledge. This immediate outcome, in turn, links to the intermediate outcomes to develop a nuanced view of wellbeing and demonstrate commitment to taking regular action to improve wellbeing. As a content specific example, the gratitude session includes encountering summaries of gratitude intervention research, discussing different ways to practise gratitude, and identifying for whom the practises work well. These activities build knowledge of gratitude practices that help to develop a nuanced view of wellbeing (e.g. interventions don’t have the same outcomes for all people) and taking action to enhance wellbeing (e.g. knowing different ways to practise gratitude supports people to take action).

Fig. 1
figure 1

Theory of Change diagram for the MARKERS program

Full size image

While the theory of change diagram provides an overview of activities in generic terms like ‘wellbeing skills practice’, a more detailed ‘content bank’ was developed. The content bank was organised by topic (e.g. gratitude, active constructive responding, workload), and contained a variety of resources for each activity for each topic. Researchers would use the content bank to select appropriate activities for each session. To ensure adaptability to specific school contexts, a broader range of content was included than we anticipated using in the program. Although content was adaptable, we ensured content included in the program covered the individual, relational, and contextual influences on wellbeing. Several PPIs were identified that met the needs to focus on the individual (e.g. gratitude interventions), and the relational (e.g. active constructive responding). Whilst research on interventions targeting contextual influences on educator wellbeing was limited, case studies of schools implementing changes to their policies and school-wide practices were provided for discussion (for example: Department for Education, 2018; Quinlan & Hone, 2020).

Theories of change also specify the indicators to be used to assess the achievement of each outcome (Reinholz & Andrews, 2020). In our theory of change diagram (Fig. 1) indicators are specified for each immediate and intermediate outcome. For example, indicators for ‘demonstrate their commitment to participating in the program’ are ‘sharing ideas in discussions, regular attendance’. Data sources for the indicators were: surveys of the wellbeing program participants before and after the program, tracking attendance to the program, facilitator observations, a focus group discussion after the program, and school wide surveys before and after the program. Formal recording of indicators was focused on the intermediate and long-term outcomes of the program. While the facilitator monitored the immediate outcomes during the sessions to provide feedback on delivery, the sessions themselves were not recorded to ensure participants could engage in discussions about wellbeing without concerns of being monitored.

3.4 Assumptions and Guiding Principles

A theory of change includes articulating the underlying assumptions regarding the process of change (Reinholz & Andrews, 2020). When summarising the change assumptions used in developing the MARKERS program, we formulated a set of guiding principles to capture the key ideas. These principles were intended to be shared with participants to act as a guide for conducting sessions (see Fig. 2). The title ‘Guiding principles for co-creating wellbeing’ reflects the belief that involving people in the change process, rather than imposing change upon them, increases the likelihood of successful change. When teachers feel heard they have more positive attitudes towards changes (Hargreaves, 2004), and experience greater wellbeing (Cann et al., 2021). These ideas align with the concept of active empathic listening, “the active and emotional involvement of a listener…that is conscious on the part of the listener but is also perceived by the speaker” (Bodie, 2011, p. 278). The phrase ‘Listen with the intent to understand, not the intent to reply’ succinctly conveys the importance of active empathic listening. The concept is particularly relevant to the MARKERS program where individuals will be discussing their own wellbeing, as it has been associated with increased psychological safety in workplaces (Castro et al., 2018). Active empathic listening includes seeking to understand others perspectives, including through asking questions (Bodie, 2011; Castro et al., 2018)—in other words an element of curiosity about the other person. We used the phrase ‘Curiosity, not judgement - about yourself as well as others’ to emphasise this, and to prompt individuals to extend the same thinking to themselves. The notion of curiosity over judgement draws upon principles from cognitive behavioural therapy and mindfulness, challenging negative self-judgments and promoting a positive and inquisitive mindset. Mindfulness and cognitive behavioural therapy have been associated with reduced stress, and greater wellbeing and personal development (Baer & Sauer, 2009; Niemiec & Lissing, 2016; van Agteren et al., 2021). The principle ‘Evidence based – research, school data, people’s experience – and challenge the evidence’ draws on the ideas of adaptive expertise and person-activity fit. Teachers are accustomed to using adaptive expertise where they utilise not only research evidence, but also student data, and personal observations to adapt their practice for optimal student outcomes within their specific context (Timperley et al., 2017). The term ‘person-activity fit’, introduced by Lyubomirsky and Layous (2013), describes how interventions are more effective for individuals when the intervention’s features (e.g., social support) align with their beliefs and needs. Using the ideas of adaptive expertise and person-activity fit, research is critiqued to identify ways in which interventions may be less effective for some people, allowing for adaptation and choice for people to focus on interventions that will have more impact for them. Additionally, drawing on adaptive expertise demonstrates an understanding of and respect for teachers’ professional expertise. Finally, ‘Appreciative inquiry lens – a strengths based positive approach to change – what do we want to create?’ encourages individuals to adopt an inquiry based and appreciative approach (as opposed to a deficit based approach), as this is valuable for promoting positive changes (Waters & White, 2015).

Fig. 2
figure 2

The guiding Principles for the MARKERS program

Full size image

4 Case Study

The aim of this study encompassed both program design and adaptation of the program in response to a particular school context. Here we present a case study that focuses on how the program was adapted to a specific school context, an intermediate school in an urban area of Aotearoa New Zealand. Intermediate schools are part of the primary education sector in Aotearoa New Zealand, and teach students aged 11 to 13 years old (school years 7 and 8). At the time of implementing the program in 2022, the school had 39 staff, including teachers, leaders, teaching assistants, and office staff. Two staff (5.1%) identified as Māori, other ethnicities included: European (59.0%), Asian (10.0%), and Pacific Peoples (5.1%). School staff had working relationships with the authors due to their school’s involvement in other research projects.

4.1 Gathering Information About Context to Inform the Adaptation of the Program

The program was designed with adaptability as a key consideration. Adaptability allows modifications to ensure the program aligns well with the school context, and meet the needs of the school community and the individual staff within the school. The first author gathered information about the case study school context through conducting two different types of interviews in early 2022. The first type of interview, ‘context interviews’, were conducted with the school principal, a deputy principal, and the wellbeing lead. The context interviews focused at the whole school level, and did not ask about individuals’ wellbeing. The second type of interview, ‘educator wellbeing interviews’, were conducted with six staff, including teachers, middle leaders (team leaders or curriculum area leaders), teaching assistants, and support staff. The educator wellbeing interviews sought an insight into the individual, relational, and school level factors influencing individuals’ wellbeing. We used the qualitative approach of framework analysis to analyse the interviews. Framework analysis uses a priori codes as well as data driven codes (Parkinson et al., 2016) allowing us to explore our predefined areas of interest in terms of barriers and enablers of wellbeing within the school. NVivo version 12 (QSR International Pty Ltd., 2018) was used to code the interviews into 29 codes across 5 categories. The final phase of the framework analysis, mapping and interpretation (Parkinson et al., 2016), resulted in the identification of four major themes. These themes informed the adaptation of the MARKERS program and are described below.

The school had dedicated time to developing its approach to student wellbeing. In 2018 and 2019 teachers engaged in professional development that focused on the PERMA (Positive Emotion, Engagement, Relationships, Meaning, and Accomplishment: Seligman, 2011) model of wellbeing. In 2020 the school purchased and implemented a positive education curriculum, where teachers devoted one hour of lessons a week to teaching about wellbeing. During the PERMA initiative teaching teams unpacked the elements of PERMA in relation to student wellbeing. One staff member estimated around 50% of current staff had experienced the PERMA initiative. However, in the educator wellbeing interviews when staff were asked about PERMA they were either unfamiliar with it, or indicated they no longer used it: “we went away from PERMA as it goes, and we started using an explicit wellbeing curriculum…that’s kind of really taken over”. The discussions of wellbeing related professional development revealed some tension between teacher and student wellbeing: “it was pointed out several times that why are we focusing on student wellbeing when we should focus on teacher wellbeing” and “certainly over recent years, we’ve really prioritised our students’ wellbeing…but that same emphasis isn’t put on staff wellbeing”.

The school had established a wellbeing team comprised of five teachers, one of whom held a formal wellbeing lead role. The team were responsible for supporting the wellbeing curriculum for students and a social club for teachers. The social club organised events such as dinners, celebrating staff birthdays, and a ‘shout out’ for staff to recognise and thank other staff. However, due to COVID restrictions these activities were infrequent over the previous two years. In the educator wellbeing interviews the staff did not mention the wellbeing team or the social club when asked about influences on their wellbeing, although they did refer to social events and social connection.

In the educator wellbeing interviews all staff referred to positive social connections within the school that supported their wellbeing. Five of the six staff mentioned their immediate team members as having a positive influence on their wellbeing, such as “I get quite a lot of positive energy from them”. However, most people highlighted the challenges of maintaining social connections. During periods of lockdown, online staff meetings were held but “people weren’t too keen on it and it felt almost forced”. When teaching resumed on site the school implemented ‘bubbles’ where a limited number of staff and students were allowed in each physical space. Approximately half of the staff commented on the negative consequences of being in bubbles, that the “segregation” was “something that’s also been difficult”, and they missed connecting with colleagues outside their bubbles. Four staff also commented that the restrictions affected the connection with senior leaders: “I feel like our leadership team has most definitely like lost touch with us as a school, like as a staff”.

One aspect of social connections that was missing was staff feeling appreciated and listened to. Previously, the social club had organised ‘shout outs’, where once a week “teachers write nice comments [about others] and pin them on the wall, peg them, and they get read out”. However, due to the COVID lockdowns and subsequent bubbles, staff were not meeting in person and the shout outs stopped. This way of regularly showing gratitude for others was not replaced by alternative organised ways of recognising staff. Half of the staff described a lack of appreciation: “there’s no praise or recognition from up top” and “you never really feel appreciated for what you’re doing”. There was some disconnect in the perception of school leadership and staff in the school in terms of being able to raise issues and concerns. Despite leaders espousing an “open door policy” and wanting staff to “be able to say whatever you want to say” some staff felt that if they raised issues there would be “no opportunity to discuss” or that they would not receive support.

One senior leader acknowledged a lack of time to discuss wellbeing as a barrier to wellbeing. Additionally, half of the staff mentioned workload as an obstacle to wellbeing. Some aspects of workload had increased due to COVID restrictions, such as managing teaching students in class and those at home isolating. The school bubbles, imposed due to COVID restrictions, reduced access to leaders and support staff, including more time-consuming communication without face-to-face interactions. Staff also highlighted a rise in the number of meetings over the previous two years, resulting in meetings after school on most days. Some staff described the effect of the resulting high workload: “my weekend is consumed by work, my evenings are consumed, no doubt my holidays will be”.

Leaders were asked about the school’s approach to tikanga Māori, to inform appropriate protocol for conducting the wellbeing program. A school-wide practice had begun for staff to incorporate pepeha, a traditional Māori greeting, when meeting someone for the first time. However, only two out of the nine interviewees mentioned the Māori concept of hauora (wellbeing). There was one other mention of tikanga Māori in relation to the beginning of year welcome for new students and staff.

4.2 Adaptations to the Program in Response to School Context

Insights obtained from the interviews were used to adapt the content and approach of the sessions to suit the staff and the school. Since the majority of staff were unfamiliar with the PERMA model of wellbeing, and only two referred to the Māori hauora model of wellbeing, we included an overview of these in the introduction session, and two more school specific wellbeing frameworks, ASPIRE (Roffey, 2017), and SEARCH (Waters & Loton, 2019). Different models were included to stimulate discussion and expose people to different viewpoints, key components in the theory of change.

Since many staff expressed the negative impact of COVID restrictions on social connections, the focus on relational influences on wellbeing in the MARKERS program fitted well with the need to rebuild social connections. A particular aspect of social interactions that staff deemed problematic was a lack of appreciation, recognition, and feeling listened to. To address the need to feel listened to, the program incorporated theory and practice of active constructive responding, which include listening attentively to others. The guiding principle “listen with the intent to understand, not the intent to reply” also aligned with the need to feel listened to. To address the need to feel appreciated and recognised we shared research on the links between showing gratitude or appreciation, and wellbeing at work (for example: Cann et al., 2021; Stocker et al., 2019; Webster, 2022). We illustrated ways to show appreciation to others in the workplace by sharing examples of ‘languages of appreciation’ in Hamrick and White’s (2020) study which provided concrete examples of actions that people can take. Languages of appreciation are based on an adaptation of ‘love languages’ which only has a limited body of research that supports the idea (for example: Hughes & Camden, 2020) and also faces some criticism (Lewandowski, 2021). However, we decided the limited and mixed evidence would provide an opportunity to critique research in line with the MARKERS guiding principle ‘challenge the evidence’. This approach helps build a nuanced understanding of wellbeing that acknowledges the limitations of wellbeing actions, as outlined in the theory of change.

In response to staff identifying workload as a barrier to wellbeing we included a session on workload reduction as part of the program’s focus on organisational influences on wellbeing. Research on the links between educator workload and wellbeing was plentiful, but there was a paucity of research on organisational interventions to reduce educator workload. Given that changes at the organizational level typically require more time to implement, involving coordination across the whole school, we designed the workload session to facilitate discussion on possible changes. As a senior leader and wellbeing leader were attending the sessions, they had the potential to continue these discussions within leadership teams, and relay feedback from staff in the sessions. To support these discussions, we used a range of resources, including case studies and resources created for schools in the UK (Department for Education, 2018) as there were no equivalent Aotearoa New Zealand based resources.

The school’s approach to tikanga Māori was taken into account when conducting the wellbeing sessions. Based on this, the first author who facilitated the program opened the first session with a karakia (traditional Māori protocol for commencing gatherings) and included a pepeha (traditional Māori protocol for introductions). Given the information provided in the interviews no further consultation was needed in order to engage with the existing tikanga at the school.

The school’s senior leaders reviewed the proposed program, resulting in adjustments to the session’s timing and duration. Given the leaders preference for face-to-face sessions, the program was scheduled to take place between June 2022 and October 2022, after the relaxation of COVID restrictions. Due to the learning priorities of the school, leaders were reluctant for regular classroom teachers to be covered by substitute teachers, therefore sessions were arranged to occur after lessons. This timing was also a factor in reducing the duration of sessions from two hours to one hour, to ensure they finished at a reasonable hour and did not add excessively to educators’ workload.

Program sessions were also adapted based on participant input. Following senior leaders review of the program, the first author contacted school staff to recruit participants. Eight staff members expressed interest, and seven of those staff members committed to participating in the program. The participants, four male and three female, had been working in the education sector for an average of 9 years. In the first session, participants were provided with an overview of the program and given the opportunity to ask questions and request adjustments to session sequence. As a result, changes were made to the order of content as participants expressed a desire to learn about stress coping strategies earlier in the program. During the sixth session, a short survey of participants was conducted to get feedback on topics they wished to recap. Session seven was adapted accordingly.

Each session followed the activities set out in the theory of change with a focus on a particular topic. The topics for each session were:

  1. 1.

    Introduction to the program. Included karakia and pepeha (traditional Māori protocol for commencing gatherings and introducing people) and gaining agreement on the topics within the program.

  2. 2.

    Wellbeing definitions and models.

  3. 3.

    Stress coping strategies, stress mindset, and the stress cycle.

  4. 4.

    Gratitude and appreciation.

  5. 5.

    Active constructive responding.

  6. 6.

    Workload in schools.

  7. 7.

    Recap session. Included a review of topics and discussion on plans going forward.

4.3 Evidence of Achieving Program Outcomes in the Theory of Change

Whilst the theory of change (Fig. 1) guided the development and adaptation of the MARKERS program, another valuable element is its use of outcomes to evaluate the program. In this section, we provide a brief assessment of each of the theory of change outcomes. The intermediate outcomes, to be achieved over the course of the program, were that participants would:

  1. 1.

    Demonstrate their commitment to participating in the program.

  2. 2.

    Develop a nuanced view of wellbeing.

  3. 3.

    Demonstrate their commitment to taking regular action to enhance their own and others wellbeing.

And the long-term outcome was:

  1. 4.

    Improved educator wellbeing

The data used to evaluate the outcomes were gathered through surveys and a focus group. All seven MARKERS participants completed a survey immediately before and after the program (May and October 2022). A week after the program six of the seven participants attended a focus group. Two surveys comprising wellbeing scales and social network questions were sent to all school staff in October 2021 and October 2022. To collect social network data respondents were presented with a roster (a list of all staff within the school) and asked “Which colleague(s) give you an increase in positive energy after an exchange? By ‘energy’ we mean feeling inspired, a positive, encouraging, and motivating interaction”. The resulting social network data then provides information on every energising interaction present within the network of school staff. We created a matched dataset for this social network which included people on the roster at both time points (i.e., excluding anyone joining or leaving the school), which resulted in a sample of 32 staff, including five of the seven MARKERS participants. Completion rates for the network questions at each time point were 87.5% and 81.3%. Data analyses included paired t-tests to check for changes over time, framework analysis of the focus group transcript to identify key themes, and Stochastic Actor Oriented Models (SAOMs) to analyse the social network data for statistically significant changes in energising interactions over time for all school staff and for the MARKERS participants. For a full account of the impact of the MARKERS program including social network analyses and qualitative data see Cann et al. (2023a).

All seven program participants attended sessions regularly (five or more of the seven sessions). A further indication of their commitment to participate, intermediate outcome 1, were survey responses that indicated they felt more comfortable discussing wellbeing with colleagues. The rating for the question “I feel comfortable discussing wellbeing with colleagues” increased from before the program (M = 4.29, SD = 1.496) to after the program (M = 4.86, SD = 0.690), although changes were not statistically significant, t(6) = -1.188, p = .280, Cohen’s d = -0.449 (medium effect size). Comments in the survey also provided evidence of this, such as “more conversations about well-being”, as well as focus group comments such as “this is the language I can use with people…how I can discuss these things with other people”.

Findings from the focus group with participants showed support for achieving intermediate outcomes 2 and 3. A nuanced view of wellbeing was evident, as participants discussed advantages and disadvantages of different models of wellbeing, and respected others’ different perspectives and preferences regarding the activities. All participants indicated a habit or change in behaviour that resulted from the program, indicating they had committed to taking regular action to enhance their wellbeing, e.g. “I do my gratitude things at the end of the week”. One of the most notable transformations was a change in the relational space. All participants indicated an increased awareness around how they interacted with others, such as actively listening, expressing gratitude and appreciation, and the vocabulary they used around wellbeing, e.g. “you’ve had an impact using appreciation in the office”, and “a bit more mindful, especially with listening when having conversations with people”, and “I think all of us have gone away and thought a little bit harder about the effect we have on others”.

Social network data from the whole school reinforced the finding about an improved relational space, with the MARKERS program participants significantly more likely to build positive relationships over time than the other educators in the school. The Stochastic Actor Oriented Models (SAOMs) analysed the changes to energising ties between school staff over time. The results showed that MARKERS program participants had a positive and statistically significant ‘ego effect’, indicating that they have a tendency to build energising ties. However, the whole school network ‘outdegree effect’ was negative and statistically significant, indicating that other school staff did not tend to build energising ties.

The MARKERS program participants showed some evidence of achieving the long-term outcome of improved educator wellbeing, but wellbeing somewhat decreased for other school staff. The measures for the satisfaction subscale of wellbeing for the MARKERS participants increased from the pre-test in May 2022 (M = 4.38, SD = 0.411) to the post-test in October 2022 (M = 4.54, SD = 0.401), although changes were not statistically significant, t(5) = 2.000, p = .102, Cohen’s d = 0.816 (large effect size). Note that given the small sample size (n = 6 for this subscale) paired t-tests can only detect Cohen’s d effect sizes above 1.44 (for ⍺ = 0.05, power = 80%, and a two-tailed test). The small increase for MARKERS participants differs to other school staff for whom the satisfaction subscale of wellbeing decreased from October 2022 (M = 4.72, SD = 0.506) to October 2021 (M = 4.47, SD = 0.887), although changes were small and not statistically significant, t(17) = 1.33, p = .200, Cohen’s d = 0.314 (small effect size). It is important to exercise caution when interpreting these findings due to the limited number of participants in the program. As a result, the study has low statistical power, and only very large effect size changes can be reliably detected. Other evidence suggests that MARKERS participants experienced an increase in their wellbeing as their ratings for the statement “this wellbeing program helped to increase my wellbeing” showed all participants either agreed (43%) or somewhat agreed (57%). Considering this alongside the statistically significant increase in their energising interactions, the MARKERS participants did experience positive effects due to the program. However, despite the MARKERS participants increase in wellbeing and the statistically significant increase in their energising interactions, this did not appear to influence their colleagues as they did not experience any increase in wellbeing or energising interactions. However, it is important to note that at the time of the survey no changes had been made to school policy or procedures.

5 Summary

Within the field of positive psychology there is a growing recognition of the importance of relational and contextual influences on wellbeing (Kern et al., 2020; Lomas et al., 2021). With a lack of whole-school approaches to wellbeing (Waters & Loton, 2019) we argue that interventions to improve educator wellbeing are in particular need of a greater focus on the relational and contextual influences on wellbeing. By outlining the design process and theory of change for the MARKERS educator wellbeing program we illustrate how individual, relational, and contextual influences on wellbeing can be addressed. An important aspect of considering contextual influences on wellbeing is that interventions include a degree of adaptation to respond to context (Halliday et al., 2019). Our case study illustrates how we tailored the MARKERS program to a particular school context by gathering comprehensive input from staff members. This allowed us to identify concerns related to the lack of recognition and appreciation, and subsequently modify the program to incorporate activities targeted at addressing these concerns (such as active constructive responding and appreciation). Following the implementation of the adapted program, noticeable improvements were observed in the relational space. Participants reported increased awareness of expressing gratitude and appreciation in their interactions with colleagues, and demonstrated improvements in their wellbeing.

The design, adaptation, and implementation of the MARKERS program offer valuable insights for schools, policy makers, and researchers interested in adopting a whole-school approach to wellbeing that encompasses individual, relational, and contextual influences on wellbeing:

  • Conduct a school-specific evaluation of needs: to effectively shape the content and delivery of a wellbeing program, it is crucial to assess the specific needs of staff and the prevailing school culture.

  • Involve school staff in decisions about adapting the program: teachers are experts at adapting practice to their particular context (Le Fevre et al., 2016) and their active involvement in decision making will foster positive attitudes towards changes, as they feel heard and valued (Hargreaves, 2004).

  • Draw on qualitative and quantitative data to evaluate the impact: statistical power may be limited by the number of staff at a school (an average of 18 teachers per school in Aotearoa New Zealand in 2021: Education Counts, 2022). Supplement quantitative data with qualitative data to capture rich descriptions of staff experiences.

  • Consider policy and other external influences: acknowledge the extent to which influences outside of the direct control of schools are impacting educator wellbeing. For example, staffing shortages or class sizes were identified as major issues by 53% of principals in 2019 (Wylie & MacDonald, 2020). The extent of external influences will vary across schools, reinforcing the requirement for a school-specific evaluation of needs.

These considerations offer valuable guidance on adapting educator wellbeing interventions to specific school contexts. Given the increasing recognition of the importance of relational and contextual influences on wellbeing, it is crucial to incorporate these factors into program design. By presenting our program design and case study, we provide an exemplar of attending to relational and contextual influences, and contribute to the discourse around developing whole-school approaches to wellbeing, in order to create positive institutions (Seligman & Csikszentmihalyi, 2000). We advocate for further research on adapting educator wellbeing interventions to different environments, in order to bridge the gap between psychological interventions and unique educational contexts.