Introduction

Promoting student wellbeing and preventing child and adolescent mental illness have become global concerns and priorities for education (O’Reilly et al., 2018). Typical efforts to support student mental health within education have adopted content approaches (e.g. teaching content through psychoeducation and curricula) and experiential approaches (e.g. creating opportunities for wellbeing practices through positive interventions), with the intention of enhancing individual students’ capabilities and opportunities to feel and function well psychosocially and academically.

Over the past decade, considerable progress in identifying content and experiential approaches to support student wellbeing has occurred (Kern and Wehmeyer, 2021; see also Owens and Waters, 2020 for a recent review of school-based PPIs). However, Allison et al. (2020) argued that wellbeing is not only fostered through targeting the knowledge and capabilities within the students themselves but also through contextual conditions. As such, a systems lens is beneficial, referring to a deliberate way of thinking to reveal the underlying elements and interconnected relationships of systems and how to ‘map, measure and understand these dynamics’ (Savigny and Adam, 2009 p. 20). In this current paper, the inquiry of research relates to the classroom as a system, given the classroom environment is known to impact student wellbeing (Gupta and Gupta, 2013; Soini et al., 2010).

Adopting a systems perspective brings numerous complexities, such as what system boundaries are necessary, who should be included, and which conditions are relevant (Kern et al., 2020; Lomas et al., 2020). Such complexity is common in systems given ‘the inter-relationship, inter-action, and inter-connectivity of elements within a system’ (Chan, 2001, p. 1). Quantifying these complex system boundaries requires a need for ‘simplexity’ — simple ways of making sense of and actioning upon a system with mindful awareness of the complexity that exists (Kern et al., 2020). Frameworks provide one type of simplexity, offering a guiding structure to describe what the boundaries of a wellbeing system may look like, making the invisible visible. Associated rubrics help to make frameworks tangible and actionable, informing appropriate interventions. In this paper, we describe the development of the Flourishing Classroom System Observation Framework and Rubric (FCSOFR), which provides an organising structure with descriptive statements for identifying observable manifestations of flourishing classrooms. The framework extends content-driven approaches that target individual capabilities for wellbeing by providing specific foci on the capabilities of members of a classroom system and the rubric informs conditions through which such capabilities are enabled or hindered.

First, the importance of student mental health and wellbeing is highlighted. Second, the classroom as a system requiring a systems-informed perspective is described. Third, the development of the FCSOFR through a Delphi study is described. Fourth, possible applications and implications of the framework for research and practice are identified. Finally, potential future directions for continuing to evolve systems-informed approaches to supporting student wellbeing in schools are suggested. As such, clarity and structure within the boundaries of a classroom system defining its observable elements for the purpose of student learning and wellbeing are provided.

Student Mental Health and Wellbeing

Mental illnesses (conditions that lead to changes in emotions, thinking, and behaviour associated with distress and problems functioning) (American Psychiatric Association, 2018) are the leading cause of disability in children and youth (Erskine et al., 2015; McGorry et al., 2014). Child and adolescent mental illnesses are particularly prevalent, harmful, persistent, and recurrent (Erskine et al., 2015; Khan, 2016). While specialist child and adolescent mental health services are central in responding to and treating mental illness, schools are well situated to work parallel to this and promote student wellbeing and prevent mental illness.

It is contended that such wellbeing-focused efforts occurring in schools worldwide need to include both content-based and context-based approaches. Establishing the importance and impact of classroom conditions for learning has been legitimised with significant contributions such as the Classroom Assessment Scoring System that measures interactions in classrooms (Pianta et al., 2008, Pianta and Hamre, 2009). Furthermore ‘social settings’ research considers how social processes and resources work as a system and can lead to improvements; however, there is limited research overall on such social settings and intervention pathway (Tseng and Seidman, 2007). We suggest then that to enhance individual-based approaches and extend the focus to additionally examine the conditions within the school and classrooms that enable, or hinder, wellbeing, a systems-informed perspective is required (Fixsen et al., 2018; Kern et al., 2020; Lomas et al., 2020).

The Classroom as a System

A system is defined as ‘an interconnected set of elements that is coherently organised in a way that achieves something’ (Meadows, 2008, p. 11). Systems can range from small to large and simple to complex. The classroom — the focus in the current study — is an example of a complex system, which impacts upon student learning and student wellbeing (Allison et al., 2020). Classrooms are a living human system, which is different from the sum of the individual parts (Kern et al., 2020). The classroom is composed of elements, including the teacher, the students, the learning environment, classroom rules, and behavioural norms, as well as the interconnections amongst these elements (e.g. how the teacher responds when a student behaves in a manner that violates a rule or expectation). Each classroom has static, explicitly and/or implicitly established structures, but also is dynamic, multifaceted, and organic, with the environment continually shifting as the different elements within the classroom act, react, and interact (Burns and Knox, 2011).

When considering what contextual elements are needed to create a flourishing classroom, Allison et al. (2020) proposed the Flourishing Classroom System Model (FCSM) as a conceptual model of relevant classroom system elements (see Fig. 1).

Fig. 1
figure 1

The Flourishing Classroom System Model (FCSM)

Full size image

The FCSM was adapted from Fish and Dane (1995) Classroom System Observation Scale, which identified communication, cohesion, and flexibility as elements of the classroom system. However, Fish and Dane’s scale is over 25 years old and focused on learning, without explicit consideration for wellbeing. The FCSM thus provides an updated version that adds a wellbeing dimension, based upon the SEARCH meta framework (Strengths, Emotional management, Attention and awareness, Relationships, Coping, Habits and goals) which was developed from a bibliometric review of 18,403 studies and action research (Waters, 2019; Waters & Loton, 2019). The current study draws upon the FCSM as a conceptual starting point for developing a practical, empirical framework and rubric for observing classroom flourishing from a systems perspective.

Observation as an Approach for Making the Invisible Visible

As described above, classroom functioning depends not only upon individual elements within the classroom, but also the liminal spaces amongst different elements (Meadows, 2008).

Tseng and Seidman (2007) through the lens of systems theory and social settings suggest, for example, elements such as social processes and organisation of resources influence setting outcomes. Observation provides an approach to capture what is directly visible and what is observable through the various senses within a classroom system. For instance, imagine a class that appears focused and quiet, and yet listening to the tone of voice of the teacher and students, it may become apparent that the order arises from compliance and fear of reprimand rather than flow and engagement. Observation captures naturally occurring, real-time behaviours within the natural environment (Johnson et al., 2016; Leff et al., 2011). Capturing the dynamics of a classroom system, conceptualising, and describing these conditions as an observable phenomenon can bring clarity and visibility to concepts that are otherwise complex and intangible (Burns and Knox, 2011; Rosas, 2017).

Adopted for diagnostic, evaluative, and counselling purposes (Leff et al., 2011; Neisworth and Bagnato, 2004), observational approaches have a long history within both education and psychology (Apter et al., 2010; Lawson, 2011). For instance, observations by teachers, parents/carers, and practitioners are often used as part of psychological assessments and evaluations to diagnose mental health, behavioural, and developmental issues (Apter et al., 2010; Neto et al., 2019; Zander et al., 2016). Within classrooms specifically, observation has been used for constructs such as pedagogical practices (McPherson et al., 2018), teacher growth (Wood et al., 2015), student engagement (O’Malley et al., 2003), technology integration (Elmendorf and Song, 2015), emotional support, resilience (Pianta and Hamre, 2009; Tapp et al., 1995), classroom behaviour management (Reinke et al., 2011), and classroom climate (Tapp et al., 1995). Direct classroom observation is regarded as both an objective and authentic method (Haep et al., 2016).

The challenge, however, is that the possibilities of what could be observed from a systems perspective could be quite expansive, meaning boundaries are practically necessary (Kern et al., 2020). Observation-based frameworks are one approach for providing structure and creating boundaries of observation, with related rubrics offering specific criteria to enable observing aspects of the system that are relevant to different areas.

Observation-based frameworks and rubrics differ in form, dependent on their purpose, underpinnings, and instructions (Praetorius and Charalambous, 2018). Thus, fitness for purpose is critical. We contend that there are three significant gaps in existing classroom observation frameworks. First, existing observation frameworks tend to focus on a single element of the classroom, such as the students or the teacher, with most existing frameworks focused on the teacher (Praetorius and Charalambous, 2018). Second, existing frameworks emphasise aspects related to learning, with little direct consideration of wellbeing-related aspects. Third, existing observation frameworks may include observable elements that can be considered wellbeing related; however, there is no observation-based framework that maps the wellbeing elements specifically into one collective framework and supporting rubric. The current study addresses these gaps by empirically developing a framework and rubric that maps the classroom as an observable interconnected wellbeing system.

The Current Study

The development of an observation framework allows teachers and researchers to clearly conceptualise and map the boundaries for wellbeing within the classroom system. The origins of the Flourishing Classroom System Observation Framework and Rubric (FCSOFR) began by drawing on the four elements of the Flourishing Classroom System Model (Allison et al., 2020) followed using the Delphi methodology to identify how these elements could tangibly be observed within classrooms.

The Delphi methodology is a rigorous, systematic, iterative process that collects the judgements of experts and translates this into a consensus, using a series of data collection and analysis techniques, interspersed with multiple rounds of feedback (Skulmoski et al., 2007). The Delphi method has been adopted broadly within the fields of both mental health and education (Björkdahl et al., 2011; Jorm, 2015) and has been used to develop existing classroom observation tools (Elmendorf and Song, 2015; Fish and Dane, 2000; Leff et al., 2011). The Delphi method was recommended by Elmendorf and Song (2015) specifically within the context of classroom observation, as the complexity of classrooms requires knowledge and experience of those with classroom expertise and experience. Further, the Delphi method is suited to meta-theoretical research (Elkington and Loter, 2013) and a useful methodology in such circumstances where a framework needs to be created (Skulmoski et al., 2007).

Method

Aligned with other Delphi studies (Cox et al., 2016), three rounds were completed independently by each expert so that participants draw on their own knowledge and were not influenced by others. All participants provided informed consent to participate in each round, and all procedures were approved by the University of Melbourne Education, Fine Arts, Music and Business Human Ethics Research Committee (no. 1955665).

Participants

To ensure that participants had significant experience in the application of both education and wellbeing science, purposive sampling was used, targeting researchers and/or educators/practitioners with expertise in wellbeing science, education, educational psychology, or positive education. The criteria to be considered an expert included the need to demonstrate formal qualifications in education and/or wellbeing, and a minimum of 5 years’ applied experience. Participants were recruited through two organisations: The education division of the International Positive Psychology Association (IPPA) and the Positive Education Schools Association (PESA). Experts known within the authors’ professional networks were also invited. Invitations were sent directly to individuals via email.

Participant attrition in Delphi research is a common issue, given the length of time between rounds resulting in their participation changing from round to round (Keeney et al., 2011). A consequence of participant attrition in a Delphi study means that the consensus results achieved can be weakened as the overall participant size reduces (McPherson et al., 2018). Given Delphi research generally has some attrition challenges, Jorm (2015) suggests that the stability of results across a Delphi panel can be achieved in panel sizes of 12 to 20 participants. Allowing for likely attrition then across the three rounds of this study, 108 research invitation emails were sent, with a copy of the plain language statement and consent form, leading to 35 participants recruited. Of these, 23 of the 35 participants completed round 2 (66%) and 18 participants completed round 3 (51%).

The experts were primarily based in Australia (n = 28), with participants also located in Germany (1), the USA (2), Singapore (2), Hong Kong (1), and Denmark (1). Of the original panel, there were seven researchers, seven teachers, five school administrators, three psychologists, three education consultants, two school senior leaders, two academics, two who were a combination of teacher and researcher, one psychology professor, one director of a national education service, and one positive education trainer. Table 1 provides an overview of the experience of the panel members, with participants able to indicate more than one background category. Further specific demographic information was not collected to maintain ethical integrity and protect the identity of the participants.

Table 1 Indicators of the expertise of panel members across the three rounds
Full size table

Procedure and Analytic Process

Across the three rounds, participants completed the requested tasks online using the Qualtrics survey software. Figure 2 summarises the Delphi process, which is described in more detail below.

Fig. 2
figure 2

Summary of the Delphi data collection and analysis process

Full size image

Round 1: Item Generation

Round 1 aimed to generate a large amount of data on the observable behaviours that indicate collective flourishing within the classroom (McPherson et al., 2018). Participants were asked open-ended questions, such as ‘How would you describe group level wellbeing in the classroom?’ and ‘What does the classroom look like as a whole when it is flourishing?’. The full questionnaire that participants were asked to complete for round 1 can be found in the Appendix.

Braun and Clarke’s (2006) approach to thematic analysis was used to identify distinctive categories, with specific statements aligning to each category. Initial coding was completed by the first author, which was then reviewed, discussed, and refined with the other authors, with categories only retained with a full consensus, but all items retained, resulting in 15 categories, aligning with 5 broader dimensions, and indicated by 254 indicators. Specifically, first there was familiarisation with the data, identifying observable statements indicative of flourishing at the group level. Statements were then coded into categories, which were determined both deductively and inductively (Roberts et al., 2019). Deductively, analyses began with the SEARCH wellbeing model, coding statements according to the six dimensions. However, to not be overly constrained by an a priori model, this was complemented by inductively identifying additional themes. Related categories were then clustered into broader dimensions, resulting in five dimensions, each consisting of three categories and a varying number of observable indicators.

Round 2: Definition and Statement Development

Although in round 1, the 254 indicators with the 15 categories were aligned, an important part of the Delphi process prioritises participant expertise over researcher expertise. Round 2 drew on a categorical analysis approach, in which responses are grouped into mutually exclusive categories (Shane and Simonoff, 2001). Presented to the participants were 254 observable indicators, asking them to allocate the items to one of the 15 categories. Responses from round 1 were randomly listed, and participants indicated which category the item was best aligned, if any. An item was determined to be part of a category if more than 50% of participants placed the item into a category, and not in others. Items that did not achieve a majority were excluded from subsequent consideration. This type of consensus is reported to be the most common form of consensus in a systematic review of Delphi studies (Diamond et al., 2014).

In addition, based on the observable indicators provided by participants, we identified and sorted items into four emerging elements of the classroom system: students, the teacher, the class as a collective whole, and the learning environment. We then generated definitions of each category, across the identified elements, based on items that through consensus aligned with the category, and aided by returning to the academic literature to further build the emerging statements. A representative name was given to each dimension and category, acknowledging that the names and alignment of categories with the broader dimensions are somewhat arbitrary, recognising that others may have chosen different labels and clusters.

Round 3: Confirmation and Refinement

Round 3 aimed to obtain consensus and feedback on the definitions developed in round 2. Participants were asked to review the definitions and statements for classroom elements and category and rate on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree) for clarity and accuracy (other Delphi studies have used 4-to-7-point Likert scales; Robinson et al., 2016; Trevelyan and Robinson, 2015). For example, participants were asked ‘The definition of student voice in a flourishing classroom is clear and accurate’. They were also asked to rate the element (student, teacher, group, learning environment) that the statement aligned with. For example, participants were asked ‘The description of strengths is clear and accurate in relation to the teacher’. Participants were also given the opportunity to provide suggestions on word changes and to provide comments on the definitions and statements: ‘Please provide any comments or feedback on the definition, the descriptions, and your ratings in regard to the (dimension and elements) of a flourishing classroom below.’

Category definitions and observable statements were restrained across the four elements when 75% or more of the participants agreed that the category definition and element observation statement were clear and accurate. The percentage for consensus can vary with Delphi research; however, 75% is reported to be the most common based on one systematic review (Diamond et al., 2014) and as such was adopted for this study. Of the three items that did not reach consensus, the feedback provided was used to make amendments to these descriptions.

Results

Round 1: Item Generation

In round 1, participants generated 254 potential observable indicators of classroom flourishing (e.g. The teacher has a warm rapport with the students, The class identifies as a team). As illustrated in Fig. 3, thematic analysis identified 15 categories, which aligned with five overarching dimensions: relationships, flexibility, communication, engagement, and mastery.

Fig. 3
figure 3

Categories and related dimensions of classroom flourishing, identified through Delphi round 1

Full size image

Round 2: Definition and Statement Development

In round 2, participants allocated the 254 indicators to the 15 categories, indicating ‘none’ if the item did not fit with any of the sub-dimensions. Thirteen items were indicated as irrelevant and were removed, resulting in 241 indicators. On average, 16 indicators were assigned to each sub-dimension (range 3 to 33). For example, ‘students seek help and support from others’ and ‘the teacher encourages students’ were classified by participants as ‘supportiveness’, which is part of the ‘Relationships’ dimension. Figure 4 illustrates the number of items categorised by participants for each category.

Fig. 4
figure 4

Number of items aligning with each category by participants in round 2

Full size image

The 241 items were also further classified into four classroom system elements: students, the teacher, the class as a collective whole, and the learning environment. Figure 5 provides a simplified visualisation of these dimensions, sub-dimensions, and elements, representing the non-linear and interconnected synergy they have with each other. It is important to recognise these sub-dimensions are interdependent with each other and the elements, building on and reinforcing each other.

Fig. 5
figure 5

Dimensions, related categories, and elements of a flourishing classroom system

Full size image

Relevant items were then combined into working definitions for each sub-dimension. The definitions were created by referring to definitions within the fields of educational psychology and positive psychology and the dictionary. The observation statements for each of the four elements were created using both the original statements and further wording clarifications from the academic literature.

Round 3: Definitions and Observable Statement Confirmation and Refinement

In round 3, participants were asked to review the definitions and behavioural observation statements, rating each on a 5-point Likert scale for clarity and accuracy. Participants were also invited to suggest wording refinements and offer additional feedback on the definitions and statements. Using 75% agreement as an indicator of agreement, the definitions for 14 of the 15 categories obtained consensus, and 58 of 60 observation statements for the students, teacher, the group, and the learning environment elements obtained consensus. Within this, qualitative comments provided suggestions for improving the wording of the definitions and statements, streamlining sentences, and removing jargon. For example, recommendations included substituting words such as ‘punishment’ to ‘consequences’, replacing the word ‘ignore’ to ‘manage distractions’ and amending ‘attention wanes’ to ‘proactively bring student attention back to the task’.

The definition for the ‘Goals and Accomplishment’ category did not reach consensus (61% consensus). The original definition was ‘Clear aims and intentions that are worked towards and met with commitment to achieving a set outcome through hard work or effort’. Participants indicated that it was wordy and somewhat confusing. Based on this feedback, the definition for Goals and Accomplishment was revised to ‘Commitment to achieving clear aims and intentions that are worked towards and met with effort’. Two of the statements for a specific element also did not achieve consensus. Within the ‘Coping’ category, the original statement for the student element ‘Students persevere in the face of challenge, are willing to struggle with a task, are flexible, adaptive, resourceful, display resilience, optimism and courage and effectively deal with stressors, learning challenges and failures’ obtained 69% consensus. Incorporating feedback, we amended the statement to ‘Students persevere in the face of challenge, are willing to struggle with tasks and use appropriate strategies to manage stress, demanding experiences and failure’.

Within the ‘Supportiveness and Belonging’ category, the learning environment element obtained 69% consensus, originally stating ‘The physical learning environment is designed to promote warmth and welcome’. Based on feedback, it was decided to provide a statement in relation to the learning environment for each overarching dimension versus the more granular category; as such, the statement was revised to a description for the learning environment element for the relationships dimension, namely ‘There is visibility of all students within the classroom. The classroom is safe, warm, inviting and designed to allow for inclusion, connection, collaboration and cooperative learning.’ In addition, although we originally combined safety and belonging in round 1, it was clear from ratings in rounds 2 and 3 that belonging and supportiveness were more closely aligned, with safety representing a separate category and we amended the framework to reflect this. Table 2 provides the final FCSOFR.

Table 2 The Flourishing Classroom System Observation Framework and Rubric
Full size table

Discussion

Schools have the potential to be at the forefront of mental illness prevention and wellbeing promotion for young people (Owens and Waters, 2020; White and Kern, 2018). A recent review of the field found that school-based mental health approaches focus at the level of individual student (skills, mindsets, characteristics) with significantly less focus on the contextual and system factors that shape student wellbeing (Waters and Loton, 2019) such as the classroom system. Yet classroom systems can be prevention-orientated contexts for students.

From a systems perspective, it is necessary to make visible what is often invisible, which includes the different elements within a system (in this case, students, teachers, the group as a collective, and the classroom environment). Increased visibility of wellbeing systems can occur through observation, in terms of both what is visible and use of other senses to capture the liminal spaces that occur amongst different elements of the system (Meadows, 2008), yet boundaries are also needed to make observation possible (Kern and Taylor, 2021; Kern et al., 2020).

The current study adopted the Delphi methodology to address these needs by developing a framework and corresponding rubric that identified the boundaries and maps the observable characteristics of a flourishing classroom system. The research began in round 1 by asking experts about collective wellbeing as a group phenomenon. What emerged however was clear independent and interdependent dimensions and elements. These dimensions and elements perhaps are not surprising, given the classroom is a social system, with all members responsible for having an influence over both stabilising and changing the system (Doll et al., 2012). The emergent framework, named the Flourishing Classroom System Observation Framework and Rubric, included five dimensions, with three sub-dimensions, manifested through four classroom system elements: the individual student, the teacher, the group as a collective, and the learning environment.

Individual Students

If wellbeing efforts in schools aim to promote positive functioning and decrease the occurrence and severity of mental illness, then young people need to develop individual capabilities to optimise their mental health, despite any challenges they may experience (Dawood, 2014). Across schools in Australia and worldwide, students have been explicitly taught varying degrees of wellbeing knowledge and behaviours. When such lessons are taught with fidelity, there is often measurable impact on both wellbeing and learning outcomes (Durlak et al., 2011; Sklad et al., 2012; Waters and Loton, 2019; White and Kern, 2018).

However, there are concerns regarding the replicability of programmes and their efficacy outside the boundaries of the research studies in which they were found to have efficacy (Hone et al., 2015), reinforcing the importance of evidence-based programmes being integrated with practice-based evidence (Cook and Cook, 2016). Furthermore, there appears to be a limitation of durability within such programmes (Dix et al., 2012) such that to foster and sustain these individual knowledge, skills, and behaviours in students, practice and feedback is needed. This FCSOFR provides a rubric for identifying students displaying higher wellbeing within the context of their classroom. Furthermore, the FCSOFR provides a map of explicit behaviour that can be practised and developed to promote individual student flourishing.

The Teacher

Although it is important to focus on students’ wellbeing, student functioning is also interconnected with and impacted by their teachers. It is not surprising then that in the context of classrooms, the teacher was identified as one of these key system elements that supported wellbeing (Graham et al., 2011). The classroom teacher has repeatedly been found to influence student wellbeing and academic outcomes (Kaya and Erdem, 2021; Ramachandram, 2016). For example, the level of altruism, grit, and patience of the teacher amongst other attributes is associated with student wellbeing (Beltman et al., 2011).

The wellbeing or illbeing of the teacher impacts on both the functioning of the teacher and outcomes for students in 22. This is demonstrated by research identifying the stress reported by teachers is correlated with higher levels of stress in their students (Schonert-Reicht, 2017). In contrast, where teachers have higher wellbeing, this is associated with higher prosocial behaviour and lower emotional and behavioural problems for students in their classrooms (Breeman et al., 2015). Teachers then have a significant role in role modelling wellbeing and reinforcing the practice of wellbeing skills in the classroom (Conley et al., 2015).

One of the key criticisms of the early applications of positive education was the overfocus on student wellbeing, bypassing the teachers’ needs (Kern and Wehmeyer, 2021). Not focusing on teacher wellbeing overlooks a key factor that can contribute or detract from wellbeing in schools and squanders a potential key wellbeing change agent within schools (Allison et al., 2020). The teacher is optimally placed to be a key influencer in providing feedback and role modelling wellbeing within the classroom.

Unfortunately, many teachers report feeling ill equipped in the space of student mental health (Reinke et al., 2011). It is also important to recognise the boundaries of the role of a teacher and not expect them to have the skills or provide the services of a mental health practitioner. The FCSOFR and specifically the descriptions of observable indicators under the element of Teacher outlined in Table 2 can provide teachers with a smorgasbord of practices to promote the wellbeing of their students and influence learning outcomes increasing their competence and confidence while remaining within the scope of their role as an educator.

The Class

The class as a collective group was also identified in this research as a system element that contributes towards a flourishing classroom. The collective class has received less attention than individual students and the teacher within extant literature, yet mood, emotions, cognitions, judgements, and behaviours can occur at the group level (Barsade, 2002). For example, within the classroom, Doll et al. (2012) found that a sense of competence was contagious across students in 22, which created a shared sense of efficacy. Allison et al. (2020) argued that there is a need to further conceptualise wellbeing as a collective phenomenon, independently and interdependently constructed (see also Kern and Taylor, 2021).

Wellbeing can be collectively constructed in the classroom by the group. For example, an emotional climate, such as the excitement for Christmas can influence group dynamics (Páez et al., 2013). Further identifying as part of a group can lead to ‘group-based emotions’ where there is an emotional response to events perceived relevant to the group as a collective (Goldenberg et al., 2014). Within the classroom, emotional engagement was found to be highest when dimensions such as group identity were present (Cooper, 2013). Yet existing wellbeing frameworks fail to capture these collective wellbeing aspects in groups, which demonstrate the systemness and interconnectedness beyond viewing this as an aggregate of individuals. By incorporating the class as a specific element, the class elements and associated describable indicators presented in Table 2 of the FCSOFR recognises the powerful influence of group wellbeing as an important entity in and of itself, alongside the individual students and the teacher. Importantly, these various sub-dimensions for the class provide greater awareness for group-based phenomenon and practices such as co-regulation, collective goals and collective identity that once made visible can be more intentionally manifested.

The Learning Environment

Finally, the physical environment of a classroom contributes towards wellbeing. The learning environment is not just the context for the teacher to teach in but is itself a phenomenon that can be considered a pedagogical tool (Cremin et al., 2006). The learning environment contributes significantly to student learning outcomes. Barrett et al. (2015) examined 153 classrooms in 27 schools, identifying seven key classroom design parameters (light, temperature, air quality, ownership, flexibility, complexity, and colour), collectively accounting for 16% of the variation in academic progress. It is noteworthy that it was the design of the individual classrooms that was found to be much more important than whole school design.

While the optimal learning environment can differ from student to student (Doll et al., 2012), it clearly contributes to the classroom wellbeing system (Burns and Knox, 2011). For example, a crowded classroom can lead to increased stress and behavioural issues as students have restricted personal space (Doll et al., 2012), whereas the colour used in a room can impact on emotions, mood, and performance (Küller et al., 2009). Thus, teachers can cater to the needs of their students through the physical environment and the spatial structure of the classroom (Cooper, 2013), with opportunities to consider physical characteristics including the choice and placement of furniture and seating arrangements, visual items on the walls, noise level, acoustics, and lighting within the room (Gettinger et al., 2012). Table 2 thus provides 5 descriptions of learning environments for each dimension that can be more conducive to wellbeing within the classroom.

A Flourishing Classroom System

Importantly, it is only when individual, social, and macro levels are targeted together that the most successful impact can occur to target upstream determinants of mental health and wellbeing (lo Moro et al., 2020). In a classroom then, it is the combination of every element of the classroom system that contributes and interacts to enable members to experience a sense of wellbeing, individually and collectively.

Conceptualising a complex human system is challenging. Any framework or rubric can be imperfect, but can still be useful (Kern et al., 2020). Elkington and Loter (2013) state that the truth in an empirical model of a system is based on the ‘ability to reduce every complex proposition down to its simple empirical referents (i.e., simple observations)’ (p. 5). Capturing what is observable in a system relates to all the senses and not just that which is visual. The use of the Delphi approach to develop this framework allowed the voices and experiences of experts to make classroom elements observable. This was critical, given resources developed for teachers need to be accepted by teachers and resonate with their views of good teaching to be embraced (Bahr and Mellor, 2016). From a practical perspective, frameworks and related rubrics need to be comprehensible and not be restrictive due to complexity, specialist language, or outdated terms (Wood et al., 2015). The feedback that arose through the Delphi process then was an essential part of the process to ensure unnecessarily complex or unclear language was not adopted.

Praetorius and Charalambous (2018) recommend that when a prototype for a framework is being developed, clarity on its purpose, as well as how it is conceptualised and operationalised, is essential. The outcome of the current study is a framework and rubric that provides a structure for considering the characteristics of wellbeing in classrooms across five dimensions: Relationships, Flexibility, Communication, Engagement, and Mastery, each representing three respective sub-dimensions. The expert data also led to the consideration for how these wellbeing sub-dimensions are present in specific identified classroom elements, namely, student, teacher, classroom/group, and the learning environment, together creating a rubric for conceptualising classroom flourishing. Given the experts’ experience and wisdom was contributed through the lens of their respective cultures, based on the geographical locations of the participants, the FCSOFR may have the most transferability to other westernised classrooms.

Implications

Teachers have long needed to manage their classroom learning environments to maximise student achievement and minimise disruptions in pursuit of student engagement in academic tasks (Doll et al., 2012; Reinke et al., 2011). Thus, it is conceivable that the teachers’ scope can and should be extended to include pedagogy and classroom management approaches that support and develop student wellbeing. Indeed, one of the recommendations of the Australian Government Productivity Report into Mental Health (Productivity Commission, 2020) is to define the role and responsibilities of teachers in relation to student wellbeing. From a systems-informed perspective, if teachers are to be responsible for student wellbeing, they not only need training and professional development regarding how wellbeing capabilities can be assessed, taught, and practised (Doll et al., 2012), but also an understanding of the classroom system conditions that cultivate the context for flourishing, the latter of which contributes to wellbeing being ‘caught’ (White and Kern, 2018). The FCSOFR provides a resource that contributes to building the capacity of teachers.

The FCSOFR supports wellbeing literacy, establishing a greater clarity and vocabulary of the classroom through the lens of a wellbeing system. Wellbeing literacy capabilities can ultimately lead to ‘intentionality for wellbeing’ where language is used intentionally to maintain or improve wellbeing (Oades et al., 2021). The FCSOFR can be used as a resource for self-reflection and group reflection, encouraging the growth and development of teachers. This is particularly important given teachers can underestimate the impact they have on students, inaccurately evaluating themselves as teachers and experiences within the classroom (Haep et al., 2016). The FCSOFR provides pathways for action, by more clearly and explicitly outlining for teachers the influence they can have on the classroom wellbeing system through a series of behaviours, previously described. Teachers possessing wellbeing system competency is important given wellbeing practices need to be emphasised as phenomenon that occurs in the community (Atkinson et al., 2017). Such clarity through the FCSOFR then has the potential to act as a compass, informing and improving educational practices.

Limitations

While the current study provides a working framework and rubric for classroom flourishing, several limitations and cautions should be considered. The participant dropout rate was almost 50% across the three rounds. While the sample is still considered large enough to be effective, given 12–20 experts from the same discipline are recommended in any study (Skulmoski et al., 2007), the attrition rate means that fewer experts contributed to the final consensus. This may have resulted in a biased perspective about the best definitions and wordings. All the participants in the study were English speaking, and while participants were represented from several countries, most were from Australia, and it is possible that this framework and rubric does not have the same relevance and appropriateness within different cultural contexts.

It is important that research into positive education and wellbeing systems continues to evolve and expand. The dimensions, sub-dimensions, and elements identified for a flourishing classroom in this study may not be exhaustive, and it is possible that there are other components of such a system that have been missed. Consensus or stability of responses through a Delphi study does not necessarily ensure correct results (Trevelyan & Robinson, 2015). Still, the Delphi method can be an important first step in establishing evidence for practice and may be useful in providing researchers with expert opinions, which could lead to future qualitative and quantitative studies (McPherson et al., 2018).

Critically, this framework is not a measurement tool, and the dimensions should not be approached purely through the perspective that more is better. The FCSOFR has been conceptualised as a map to help quantify classroom flourishing by defining and describing its system characteristics, understanding that the interdependence of these characteristics can ultimately influence each other in powerful ways.

Future Research

Expanding on this study, there would be merit in students themselves being participants in what they report as observable within a flourishing classroom. The voice of students is particularly critical given they have previously demonstrated a different view of wellbeing compared to adults (Bharara et al., 2019), and as such, their perspectives on classroom flourishing could enhance this framework. There is also value in validating this framework though further research, for instance exploring if the framework has inter-rater reliability between observers or the correlation it has with student self-reported subjective wellbeing and academic achievement. There is also potential for this framework to be adopted to create a measurement tool with associated processes and applications to build on this work.

Conclusion

To effectively implement strategies that promote the wellbeing of young people within education, a systems-based approach with multiple action pathways to effect change is required. When considering how this could be applied within the school classroom, the current study adopted the Delphi methodology to create an observation framework of a flourishing classroom system with a related rubric. Consensus from researchers and experienced educators identified five dimensions with fifteen observable sub-dimensions of a flourishing classroom relating to four classroom system elements. This framework and rubric makes tangible the wellbeing dimensions, sub-dimensions, and elements of a flourishing classroom system, providing a guiding structural map to drive system-level wellbeing change. As a result of this framework, educators are strongly placed to engineer and influence their classrooms as a system to promote the wellbeing of all, including their own.