Teachers’ action research as a case of social learning: exploring learning in between research and school practice

ABSTRACT

Although action research has a history of bridging gaps between research and school practice, challenges emerge when aligning a scientific approach with development work in schools and in collaboration between research and school practice. In this article, we aim to deepen the knowledge on teachers’ action research as social learning in collaboration with a research-based PD leader. In this study, we follow two teachers conducting action research in a Swedish upper secondary school, in collaboration with a professional development (PD) leader who is also a doctoral student. Qualitative data have been used to write value-creation stories that illustrate the learning process throughout their collaboration. Our findings show how joint work between the teachers and the PD leader functions as a boundary process where two sets of practices – classroom and academic – are coordinated and contribute to the learning process.

KEYWORDS:

• Action research
• professional development
• social learning
• learning at boundaries
• boundary objects
• brokering

Introduction

Studies on teachers’ action research have shown how teachers learn collectively by developing teaching and scientific skills together, thus bridging gaps between research and school practice (cf. Bergmark, 2020; Johannesson, 2020; Ponte, 2002; Ponte et al., 2004). Additionally, we know that learning is nurtured through collaboration with other partners (Bergmark & Erixon, 2020; Olin et al., 2016). Previous research has problematised aspects such as power relations, epistemologies and changes that might occur in partnerships where teachers collaborate with different partners (Bevins & Price, 2014; Bruce et al., 2011; Somekh, 1994). However, there is a need to better understand how the process of learning emerges and is affected by the different partners involved in the collaboration.

The Swedish Education Act, 800 (SFS [Swedish Code of Statutes], 2010), which states that all education in Sweden should rest on science and proven experience, has led to increased demands on schools to undertake research-based activities and apply scientific methods to their development work. Bergmark and Hansson (2021) show that teachers and principals find it difficult to interpret the policy and struggle to enact it. One way to do it could be action research, yet a difficulty that has been identified in action research among teachers lies in mastering different sets of skills simultaneously to be able to both teach and conduct research in one’s own practice (Ponte, 2002; Wennergren, 2016). Åman and Kroksmark (2018) argue that continuing professional development initiatives must include scientific training to support teachers’ scientific work because schools in Sweden lack scientific organisation and history.

Alongside the Swedish Education Act (2010), a career reform was implemented with the intent to raise the status of teachers and foster professional development on local sites through the so-called first teachers. Although some researchers (e.g., Hardy & Rönnerman, 2019) argue that the reform rather led to reduced professionalism among teachers, Bergmark (2020) describes how first teachers or ‘lead teachers’ in Sweden could play an important role in supporting professional learning without collaborating with researchers, contributing with knowledge and skills beyond ordinary teachers’ knowledge base.

In this study, we follow two teachers conducting action research in an upper secondary school in Sweden, in collaboration with other teachers and a professional development (PD) leader.¹ In this context, teacher learning in situ as teachers develop their classroom practices through action research can be explored. Theoretically, a framework about value creation (Wenger-Trayner et al., 2017; Wenger-Trayner & Wenger-Trayner, 2020) is used to describe and understand teachers’ action research as social learning. With this framework, the values enacted and expressed by the participants come into focus and allow us to create narratives about the learning trajectories that occur in practice. Additionally, Wenger’s (1998, 2000) concepts of boundary objects and brokering help explain how different participants engaged in the collaboration contribute to the learning trajectories.

Our aim is to deepen the knowledge on teachers’ action research as social learning in collaboration with a research-based PD leader. Our research questions are as follows:

1. What are the critical aspects of teachers’ action research as a social learning process undertaken together with a PD leader?

2. How do boundary objects and brokering contribute to that process?

Action research and collaboration in professional learning

Action research has a long tradition of bridging gaps between research and practice development and has witnessed many iterations since its original forms in the 1950s, including the teacher-as-researcher movement in educational settings (cf. Zeichner & Noffke, 2001). However, several critical aspects have been identified when teachers engage in action research. One critical aspect is that action research is seldom initiated by teachers themselves but requires external incitement (Johnston, 1994). Johnston argues that the systematic, collaborative and critical process of action research “is not compatible with the way many teachers explore and understand their practice” (1994, p. 46).

Another critical aspect of professional learning through action research has been investigated by Ponte (2002), who illuminates the complexity of the required areas of knowledge and what teachers do to develop this knowledge. To further investigate this complexity, Ponte et al. (2004) epistemologically categorise teachers’ professional knowledge production through action research and distinguish among three domains of knowledge: ideological domain (understandings of norms and values), empirical domain (concerned with insights into educational reality) and technological domain (understanding of methods, techniques and strategies). The authors note that when left to themselves, teachers develop knowledge in the technical domain, but with the support of facilitators, they also develop knowledge in the empirical and the ideological domains. Based on this conclusion, they “assume that daily practice tempts teachers to seek immediate, technical solutions. Conceptual skills […] risk disappearing into the background” (Ponte et al., 2004, p. 587). When teachers engage in a programme with academics to learn through action research, the authors also suggest a shift in the facilitation model from a traditional transfer model to a more cyclical and process-oriented one.

To facilitate the resolution of the issues raised so far, collaboration with and support from researchers and critical friends have been suggested, which in turn pose difficulties in overarching power relations and differences in epistemologies and in what counts as valuable knowledge. These are well-studied areas (Aspfors et al., 2015; Bevins & Price, 2014; Bruce et al., 2011; Olin et al., 2021; Somekh, 1994), and the similarities stressed within this research are the importance of creating arenas for communication, the recognition of one another’s competencies and contributions, and sustainability. While some research focuses on the approach and the models of collaboration (Bevins & Price, 2014), other studies concentrate on the researchers’ roles as facilitators and negotiators between competencies and discourse (Bruce et al., 2011; Olin et al., 2016; Somekh, 1994). In her critical analysis of collaborative action research, Somekh (1994, p. 378) concludes, “Discourse is the most problematic issue in school-university collaboration because […] it is rooted in the other larger educational issues of what ‘counts’ as knowledge, and how we decide that knowledge is sufficiently trustworthy for us to act on it.”

In a more recent study, Hartmann and Decristian (2018) have examined the kinds of settings where teachers and educational researchers try to overcome such professional boundaries between educational research and school practice. They conclude that “[t]he decision for teachers and researchers to engage in brokering at the boundary of educational research and school practice is likely to be influenced by their professional and personal histories as well as by the context in which they operate” (p. 124). Since the authors could identify brokering activities between teachers and educational researchers in research projects in schools, network activities and professional development, they propose that brokering could reduce the gap between practice and research. However, brokering activities are complex and intertwined with the identities of the professionals involved in such activities. For this reason, the authors point to the importance of regarding this set of activities as a potential knowledge exchange and not as a venue for one practice to have an impact on another.

The theoretical framework used in this current study provides the opportunity to explore such knowledge exchange because it brings to the fore experience of practice, which “is often marginalized or assumed to be a receiver rather than a creator of knowledge” (Wenger-Trayner et al., 2017). Hence, rather than describing differences between academic research and teacher research (Sachs, 1997), our study contributes with a social learning perspective on action research and explores professional learning at the boundaries in between research and school practice, where the value of the process is not predetermined. In this way, the issue of what counts as knowledge, trustworthy enough to be acted on (cf. Somekh, 1994), is further examined.

Theoretical framework

To further explore and deepen the knowledge on teachers’ action research as social learning in collaboration with a research-based development leader, parts of the conceptual framework developed by Wenger-Trayner and colleagues (Wenger, 1998, 2000; Wenger-Trayner et al., 2015; Wenger-Trayner et al., 2017; Wenger-Trayner & Wenger-Trayner, 2020) are used to describe and understand the social learning system at play. This framework focuses on social learning, where learning is defined as an interplay between social competence and personal experience through participation in communities of practice (CoPs) and the creation of identities in relation to these communities (Wenger, 2000). A CoP can be described as having three dimensions: shared repertoire, mutual engagement and joint domain; moreover, three modes of identification – engagement, imagination and alignment – are required to become a full member of a social learning community (Wenger, 1998; Wenger-Trayner et al., 2015).

These modes of identification emphasise the importance of participation in social practices, where (from a practice perspective) “knowledge is conceived largely as a form of mastery that is expressed in the capacity to carry out a social and material activity” (Nicolini, 2012, p. 5). The social dimension of a CoP – mutual engagement – has been further elaborated and “theorized as value creation in social learning spaces” (Wenger-Trayner & Wenger-Trayner, 2020, p. 6). The authors refrain from defining what counts as value; rather, they focus on the creation of value in what they call an “operationalized view of the nature of mutual engagement in social learning” (Wenger-Trayner & Wenger-Trayner, 2020, p. 5). The notion of “social learning spaces” allows us to study social learning processes where knowledge and competence from multiple CoPs can be found within a specific social structure in a CoP.

To understand social learning systems, Wenger (2000) proposes three structuring elements, defined as 1) CoPs, 2) identities shaped by participation in CoPs and 3) boundary processes between communities. Boundaries between CoPs can be used to identify differences in ways of working in CoPs and to serve as bridges between them. Boundary objects are characterised by their ability to enable communication and coordination, as well as align activities between practices, not necessarily forcing consensus (Kubiak et al., 2015). On the contrary, the bridging of practices can be assessed along the dimensions of coordination (enabled joint activity), transparency (understanding of one another’s practices) and negotiability, that is, whether one another’s competencies and perspectives are mutually recognised (Wenger, 2000), without forcing consensus (cf. Star, 2010). Since competence and knowing within this framework are defined by the members of a community, the process of crossing boundaries can be problematic. In educational settings, for instance, teachers can regard academic competence and theorising as “an esoteric ‘ivory tower’ enterprise unrelated to their professional needs” (Carr, 2005) and thus exclude such competencies. This leads to the concept of brokering.

Brokering is a process where people can “introduce elements of one practice into another” (Wenger, 2000, p. 235) and thus create bridges between different CoPs. This is a delicate process where people who function as brokers need to have “enough legitimacy to be listened to and enough distance to bring something really new” (Wenger, 2000, p. 236). Wenger (2000, pp. 235–236) suggests various forms of brokering, such as the following:

boundary spanners: taking care of one specific boundary over time; roamers: going from place to place, creating connections, moving knowledge; outposts: bringing back news from the forefront, exploring new territories; pairs: often brokering is done through a personal relationship between two people from different communities and it is really the relationship that acts as a brokering device.

Regardless of the form of brokering, the core elements in brokers’ work are “building connections between different practices, introducing practices from elsewhere and facilitating cross-boundary experiences” (Kubiak et al., 2015, p. 81).

In this study, the notion of a CoP and the creation of identities in relation to CoPs are used as the background. The collaboration and the value-creation processes are foregrounded.

Design and method

The context of the study

The research design of this study can be described as a case study involving an upper secondary school in Sweden with an approach to school development through action research. Founded in 2014 as an independent school, from the start, it created an organisation to support professional development, including the appointment of a PD leader. The teachers in this school attend weekly meetings (the so-called learning groups) where they, supported by the PD leader, work together using an action research approach – best described as classroom action research (Kemmis et al., 2014a) – to improve teaching practices. Using Wenger’s (2000) terminology, we consider this a cross-disciplinary project, where bodies of knowledge from two sets of practices – one of teaching and the other of action research – are combined to improve classroom practices.

The first author’s double role, as the PD leader in the school and a doctoral student, involved facilitating the process, acting as a critical friend and studying the process, which included generating empirical evidence throughout the process. Hence, this study is viewed as both first- and second-order action research (cf. Feldman, 2020). It contains examples of teachers’ and the PD leader’s collaboration in teachers’ action research (first order). At the same time, it is the study of their collaboration and doing of action research (second order). Thus, the study contributes with knowledge about the conditions that either facilitate or obstruct learning in this context.

Data and analysis

To explore action research as a case of social learning, the data have been selected through the abductive approach of combining theoretical concepts with the first author’s knowledge as a researcher from the inside (Kaukko et al., 2020). The evidence has been selected from a larger dataset, generated throughout the academic year 2017/2018, to be able to write value-creation stories (Wenger-Trayner et al., 2017; Wenger-Trayner & Wenger-Trayner, 2020). The value-creation framework is a theoretical elaboration on the concept of mutual engagement and is also suggested (Wenger-Trayner & Wenger-Trayner, 2020) as a method of evaluating social learning. The framework includes seven steps, where each step either creates or does not create value. Wenger-Trayner et al. (2017, pp. 3–4) describe the value-creation cycles as follows:

Our framework includes seven value-creation cycles: (a) engaging in a social learning interaction can produce immediate value, such as enjoying the company of like-minded people or doing something exciting; (b) this engagement can generate potential value such as insights, connections, or resources; (c) drawing on these insights, connections, or resources to change one’s practice requires much creativity and learning, and thus, is viewed as generating applied value; (d) to the extent that changes in practice make a difference to what matters, social learning produces realized value; (e) if it transforms people’s identities or the broader environment, we say that it has produced transformative value; (f) successful communities engage with relevant stakeholders to ensure that their learning makes a difference: the quality of these conversations and relationships produces strategic value; and (g) getting better at supporting or enabling social learning produces enabling value.

In our analysis, the value-creation framework has been adapted and integrated with the action research process to describe the latter as a case of social learning and to identify critical aspects of the collaboration throughout the process (see Figure 1).

Figure 1. An adaptation of the value-creation framework in alignment with the action research process.

While the strategic and the enabling values stated in Figure 1 illustrate the school’s intentions and expected outcomes regarding the action research approach, the process – from creating immediate value to producing transformative value – is described through two teachers’ (Julia’s and Björn’s) first-order action research. The two examples of teachers going through all cycles were selected based on the required amount of data that should be described in order to ⁣⁣generate a rich description through “crystallization” (Tracy, 2010, p. 844). In this case, the concept means combining different methods and data to generate a deeper understanding of what is being studied. These examples were chosen because they represented two different learning trajectories among the teachers in the school. Further, not necessarily success stories – of teachers solving problems in a predictable and uncomplicated way – they responded well to the call for more realistic reports of practitioners’ experiences that would show some of the messiness throughout the process (cf. Boyle et al., 2022). The examples shared some similarities; both teachers participated in the same learning group and had been working in the school for the same amount of time and therefore shared experiences from previous development work. To explore brokering when facilitating the process, a third story from the PD leader’s perspective was created.

The stories and the analysis are based on data from the PD leader/researcher log, transcribed extracts from audio recordings from the learning groups, video recordings of the two teachers’ presentations of their action research and empirical evidence generated by the teachers (see Table 1). Combining these data enabled the writing of value-creation stories. The teachers’ names in the stories are pseudonyms. All participants were informed orally about the study and signed a letter of consent, with the right to withdraw from the study at any time. Special consideration in compliance with good research practice (Swedish Research Council, 2017) was taken for data involving students.

Table 1. Data generated by the teachers and the PD leader/researcher that have been used to write the value-creation stories.

Teachers	PD leader/researcher
Julia’s log Data from student surveys Worksheets (compiled data and categorisation sheets) Four video-recorded interviews with students Björn’s action research report	Researcher log Two audio-recorded PD meetings on analytical conversations Video-recorded presentations of Julia’s and Björn’s action research

The analytical approach can be described as follows: First, a matrix was constructed to identify the boundary objects and brokering activities available in each example. These were selected on the premises of coordination, transparency and negotiability (Wenger, 2000), that is, artefacts and ways of working that enabled joint activity (see Figure 5), which in turn allowed the process to proceed to the next phase of the value-creation cycle. Second, the different data were compiled into value-creation stories by the researcher (first author) and later validated through member reflections (Tracy, 2010), which in this case meant that the teachers read and validated the stories.

Turning to the next section, this means that the findings presented are based on the researchers’ analysis of the value creation taking place through the teachers’ and the PD leader’s collaboration during the action research processes. This approach was chosen with the perspective that the “use of value-creation stories presupposes the legitimacy of the voice of practice as a source of evaluation data and learning loops” (Wenger-Trayner et al., 2017, p. 335). Hence, to explore action research as social learning, we (the researchers and authors of this article) have drawn on the voices of the teachers’ and the PD leader’s experiences of practice, which constitute valued knowledge for researchers doing action research with teachers.

Findings: value-creation stories of learning at boundaries

The value-creation stories (Figures 2–4) are presented in this section. Each story is told chronologically, following each step of the action research process. Julia’s story (Figure 2) is that of a teacher of the Swedish-language subject who wants to increase students’ focus on learning tasks during her lessons, while working with the rhetorical process and teaching the students how to write and deliver speeches. The second is the story of Björn (Figure 3), an art teacher who designs a course map to enhance student motivation and artistic skills simultaneously. Third, the PD leader’s story (hereafter, referred to as Peter’s story) sheds light on the PD practice and the facilitation of aligning a scientific approach with the development of classroom practices. Each story starts with a summary, followed by a value-creation story narrated in the first-person point of view (Figures 2–4), and then by analytical paragraphs, highlighting the aspects that enabled the processes.

Figure 2. Julia’s value-creation story.

Figure 3. Björn’s value-creation story.

Figure 4. Peter’s value-creation story.

Julia’s story

Julia’s action research took its starting point in the school’s mutual development area on classroom management and in previous development work. She wanted to increase her students’ time-on-task (a concept that the school’s teachers had picked up previously from research on teacher efficiency; Muijs et al., 2014). She asked herself these questions: How much time-on-task do students spend during my lessons, and what do they have to say about that? The actions included structuring lessons based on local policy and research on teacher efficiency (Muijs et al., 2014). After every lesson, the students were asked to rate their time-on-task and write comments about their participation. Julia herself kept a log where she wrote reflections after each lesson. In the second step, after analysing her students’ comments and summarising their time-on-task, she planned to report back the results to the students and ask them for help regarding what she could do to increase their time-on-task.

When Julia defined her problem, she recognised an immediate value in the development area on classroom management that had been done earlier in the school in relation to her own classroom practice (see Figure 2). In her planning, research on teacher efficacy, adapted to the school’s lesson policy (boundary object), gave her ideas that she wanted to try out because she identified potential value if this would affect her students in the direction she hoped for, meaning that they would utilise the lessons and write and present their speeches on time. In Julia’s case, applied value was generated through her ability to choose methods that generated empirical evidence without drawing too much engagement away from their classroom work. Thus, her log and the students’ ratings and comments functioned as boundary objects that enabled coordinated action between classroom practice and PD practice when proceeding to the analytical phase.

During this phase, Julia asked a researcher whom she knew (brokering through pairs) and who had competencies in quantitative data to compile her students’ ratings and compare them with hers. Through a spreadsheet (a boundary object), she also asked for the PD leader’s help in categorising her students’ comments on what had helped them to accomplish their assignments. The PD leader contributed with the concepts of sayings, doings and relatings (boundary objects) from the theory of practice architectures (cf. Kemmis et al., 2014) to identify aspects that either enabled or constrained student learning. This joint process created realised value. Although Julia was unable to solve the problem and increase her students’ time-on-task or finish her action research report (see Figure 2), transformative value was created with the help of her log and empirical evidence from the students (boundary objects). In other words, the material created throughout the process enabled coordinated activity among Julia, her friend outside of school and Peter, which helped Julia gain insights about her practice and the students’ opinions on their engagement during her lessons. In this way, Julia’s story illustrates the interplay between carrying out a social activity and a material activity.

Björn’s story

Björn’s action research project was guided by this question: Can students’ motivation be enhanced if they can create meaning and orient themselves with the help of a course map? The actions included designing a course map with course elements and examination assignments that would provide students with structure during lessons and allow them to meet the course requirements. Björn also introduced the daily standup, an activity in which each student started every lesson by describing where on the course map they were and what the next step would be. To answer his question, Björn chose to create data using video-recorded interviews with his students, and for his analysis, he used a theory of motivation (Ford, 1992).

Björn perceived immediate value in combining issues from his course evaluations (boundary objects) and his interest in the students’ motivation when defining his problem (see Figure 3). In his planning, he recognised potential value if he could design a course map that would meet the students’ need for structure; at the same time, they would improve their skills in art and meet the course requirements. His plan to evaluate the attempt through video-recorded interviews (boundary objects) would also offer him insights on student motivation. Introducing the ‘daily standup’ and the course map (boundary objects) created applied value, allowing the students to work on their assignments and Björn to video record his interviews with the students about the work. Realised value was created through the joint empirical work and analyses in the learning group. Watching the video-recorded interviews (boundary objects) and trying to understand the student responses, Björn, another arts teacher in the learning group and the PD leader tried out multiple theories of learning. Through his affiliation in his research environment (brokering through pairs), the PD leader gave Björn a thesis (Blomgren, 2016) (a boundary object) that focused on student motivation and included a theoretical framework (Ford, 1992) that was transparent enough for Björn to adapt and use for his analysis – that is, he could interpret the theory and recognise how it could be useful. Insights on what motivated his students and the diversity among them generated transformative value, making Björn interested in student identity building and their learning outside of school (see Figure 3).

Peter’s story

While Julia’s and Björn’s stories are examples of first-order action research, Peter tells the story about supporting his peers in their action research processes while arranging for collaboration within the PD practice and making all participants work in alignment with locally identified development areas. His actions included trying out different tools for collaboration and negotiating his knowledge about research and knowledgeability of different theories with local history and practice.

As the starting point in the action research process is to define one’s own problem, Peter used tools, such as mind maps and interviews (boundary objects), in the beginning of the academic year to enable collaboration and opportunities to negotiate on local development areas with individual classroom practices (see Figure 4). The intention was to create immediate value so that the teachers would recognise the value in adding a scientific approach to make a difference in something they cared about. In the planning phase, to create potential value, Peter brokered between academic and local practices when helping his peers with action research designs that would have the capacity to generate empirical evidence while being doable. This would consequently create applied value – that is, they would carry out their plans. In the next phase, to facilitate the generation of realised value, Peter engaged with the empirical evidence (boundary objects) produced by his colleagues. Together, they negotiated (brokering) on different theoretical perspectives (boundary objects) that contributed to their understandings of classroom practices.

In Julia’s case, they used the theory of practice architectures (Kemmis et al., 2014). In Björn’s case, after trying out several theories of learning, they ended up with a theory of motivation (Ford, 1992) through Peter’s connection to his research environment (brokering through pairs). As the PD leader and a critical friend, in the last phase, Peter tried to facilitate the creation of transformative value through individual feedback on his colleagues’ action research and by putting it in a broader context. Introducing the sharing day, he highlighted the local development areas and the contributions that were being presented to meet the target goals for these areas. After the sharing day, he summarised the contributions and the exit tickets containing statements about what they had learned throughout the academic year and what they believed they had developed as a school.

The learning group as a boundary practice: bridging competing values

To summarise, stories about social learning at the boundaries in between research and school practice unfold within the learning group in a specific school. The boundary objects and brokering that have been identified (Figure 5) illuminate how Julia and Björn bring artefacts from their classroom practices into the learning group. In turn, Peter brings artefacts from academic practices and research to facilitate the action research, including empirical work and analyses.

Figure 5. Boundary objects and brokering activities throughout the action research process.

In beginning their processes, the creation of immediate value was dependent on the two teachers’ and the PD leader’s ability to negotiate meaning among three competing sets of norms and values: the local development areas, individual classroom practices and the action research approach. Without this mutual engagement, the process would have stopped, preventing them from advancing to the next phase of planning their actions. However, immediate value was created somewhat differently. While Julia recognised value in the local development area and research on classroom management, Björn took his starting point in course evaluations and in his personal interest in student motivation. Here, if the PD leader had insisted on Björn working with something more closely aligned with the local development area (as Julia did), moving on to create potential value would have become more difficult for Björn.

In the planning phase, potential value was created through their collaboration in formulating research questions that, when answered, had the potential to inform and solve classroom problems. Additionally, their plans had to be made with the potential to align actions for classroom improvement with the scientific approach to simultaneously create empirical evidence to be able to evaluate actions and create applied value. Without the teachers’ adaptations of scientific methods, again, the processes would have stopped, without the teachers producing empirical evidence in favour of making their classroom changes (which would have hindered their collaboration in the analytical phase) or with too much attention being paid to the scientific norms and values (which would have decreased their engagement in students’ learning).

In the analytical phase, the empirical evidence generated by Julia and Björn functioned as boundary objects, allowing them and Peter to interact and engage in categorisation and analysis to generate realised value. Despite Peter’s insights regarding local practice as a fellow teacher in the school, he would have been unable to engage in the analyses without the evidence generated by his peers. Furthermore, the negotiation on competing values continued in this phase, illustrated through the dismissal of some theoretical perspectives and the adaptations of others. Again, as pointed out in Julia’s and Björn’s stories, this exemplifies the interplay between the material and the social activities in the learning group.

In the final phase of presenting and sharing results, Julia and Björn offered their new insights (transformative value) through the action research processes. In Julia’s case, although she did not complete her report and the final step in her action research plan, the mutual engagement in the learning group enabled her to present her unfinished work without losing legitimacy in the eyes of her peers because she was still successful in her work with her students. Here, the value of Julia’s action research was not defined by her ability to solve her problem or to meet the local target goals. Rather, the value was determined on the basis of the enabling value in gaining insights about the problem and the opportunity to ask new questions and to engage in a new value-creation process. Furthermore, Julia’s case illustrated that even though she found value in planning her actions based on research on teacher efficacy, this knowledge did not ‘fix the problem’ because the students did not consider a low time-on-task a problem.

Discussion and conclusion

This study has aimed to deepen the knowledge on teachers’ action research as social learning in collaboration with a research-based PD leader. Analytical tools such as the value-creation framework and other parts from Wenger’s (1998, p. 2000) social learning theory have been used to explore 1) critical aspects of teachers’ action research as a social learning process undertaken together with a PD leader, and 2) how boundary objects and brokering contribute to that process.

To answer the first research question, we highlight three critical aspects in the social learning process. First, during the first value-creation cycle, one critical aspect entailed the negotiations on three competing sets of norms and values: the school’s local development area on classroom management, the action research approach and the teachers’ individual values in relation to classroom practice. If the school’s approach to meeting the target goals for the development area had required all teachers to do the same thing in their respective classrooms – or if the PD leader had decided on the research questions – the emancipatory aspects of action research would have been lost, with less opportunities to create immediate value, leading to decreased professionalism (cf. Hardy & Rönnerman, 2019). Moreover, a strict commitment to scientific norms and values would have hindered the creation of potential value, with the teachers being unable to recognise how they can research practice and simultaneously teach their classes.

The second critical aspect throughout the process involved the adaptations of scientific methods. For instance, from an academic perspective, Julia’s method of gathering her students’ ratings of their time-on-task, as well as her own approximations of the same, could have been dismissed as failing to meet the standards of reliability and validity. Nonetheless, the adaptation allowed the creation of empirical evidence, which in turn made the study participants’ social learning process continue towards collaboration in categorisations, which helped Julia gain insights into her practice that made sense to her. Bruce et al. (2011) make a similar observation of adaptations regarding data collection and analysis in action research. They report that for “the university researchers, the greatest learning stemmed from our need to adapt and apply data collection and analysis strategies that were manageable given the other professional demands made on teachers and given the range of research questions being asked” (p. 443).

The range of research questions raised the final critical point. As illustrated in this article, the creation of realised and transformative values depended on the broad knowledgeability of (and access to) a variety of analytical tools and theoretical perspectives to be used in the empirical work. This is a crucial point if the aim is to support teachers in developing conceptual knowledge (cf. Ponte et al., 2004) without losing the emancipatory aspects of action research, for instance, by deciding on the research questions for the teachers.

To answer our second research question, the findings illustrated how boundary objects (Figure 5) supported the connection between research and school practice and enabled joint action between the teachers and the PD leader. While the PD leader functioned as a broker between local practice and research practice, this study illustrated in particular how the collaboration in the learning group functioned as a boundary process where two sets of practices (classroom and academic) coordinated and contributed to the study participants’ social learning within the PD practice, bridging gaps between research and school practice (cf. Hartmann & Decristian, 2018).

More specifically, material boundary objects – in the form of theoretical perspectives, research articles and the evidence gathered by the teachers – made it possible to perform social activities, such as joint categorisations and analyses of the student data (coordination and joint activity). The adaptation and negotiation of theories contributed to new understandings about student learning (transparency and negotiability). This was specifically evident in Björn’s story. While the initial attempts to apply a learning theory to analyse student interviews failed because the theories seemed too grand for Björn, the theory of motivation was transparent enough for him to interpret and find useful when negotiating on how it would help him understand the students’ statements in the interviews. Finally, aligning the action research approach with the development of classroom practices was dependent on the participants’ ability to negotiate values in relation to the school’s development area on classroom management.

Hence, viewing teacher action research as a case of value creation in social learning spaces opens further discussions on action research as a PD programme and on the practical implications when implementing action research as such. Wenger-Trayner and Wenger-Trayner (2020, p. 68) stress that the social learning theory does not specify what counts as value; instead, they focus on value creation:

The centrality of agency shifts the focus from something being learned to the lived experience of value for people who care to make a difference – big or small, in the world or in themselves. The theory does not prescribe what that difference should be, only the focus on meaningfulness and agency.

Adopting this stance, for action research implemented as a method for PD to be sustainable, participants should be given recurrent opportunities to define values themselves and develop their agency. From the social learning perspective, supporting and facilitating teachers’ action research imply a focus on agency and the emancipatory dimensions of action research. Although the epistemological value (cf. Ponte et al., 2004) in teachers’ action research may differ, when facilitating the process, there must be sensitivity towards the values that teachers want to create for themselves and their students. For example, newly qualified teachers and newcomers to the profession might find value in developing their pedagogical skills and delivery of curricula (technological domain), while more experienced teachers might define values by developing knowledge in the empirical or the ideological domain.

Additionally, rather than focusing solely on scientific training (cf. Åman & Kroksmark, 2018) and doing research ‘by the book’, emphasis should be placed on generating empirical evidence to evaluate improvement efforts in a way that is doable in practice without drawing too much engagement from classroom practice and student learning. This view does not imply a laissez-faire approach to research and academic practices but suggests a mutual openness to deal with differences and “commitment to suspend judgment in order to see the competence of a community in its terms” (Wenger, 2000, p. 233). Employing the value-creation framework, as tried out in this study, can bring to the surface some issues on what is valuable knowledge to act on (cf. Somekh, 1994) and thus be an effective tool for mutual understanding that can support collaboration in professional learning between research practice and school practice.

In conclusion, viewing action research as a case of social learning entails creating personal experiences in social interplay and through participation in CoPs. Consequently, for schools that struggle to enact the policy of working on a scientific foundation (cf. Bergmark & Hansson, 2021), one way to ease the struggle is to consider PD through action research, not as a group of teacher researchers making generalisable knowledge claims, but as a group of learning partners creating values that make a difference to themselves and their students. The focus on creating experiences in collaboration with others also agrees with the invocation: “to go beyond action research as a project, we need to see it as a never-ending process, and reporting of action research should then be about where the process is at that time” (Boyle et al., 2022, p. 8). This point of view is also beneficial in terms of avoiding both a ‘celebratory stance’ in action research (Mockler & Groundwater-Smith, 2015) and the focus on solutions and ‘what works for whom’, an issue of power (cf. Wenger-Trayner et al., 2017) that if left unresolved, decreases teachers’ opportunities to develop their agency.

However, from an organisational perspective, for scientific knowledge and competence to play important roles in teachers’ professional learning, these cannot be dependent on personal relationships (brokering through pairs) with academics or fixed-term projects with researchers. Rather, to address the lack of scientific organisation and history in Swedish schools (cf. Åman & Kroksmark, 2018), these should form an integrated part of the organisation. Nonetheless, knowledge and competence in how to work scientifically, as defined socially by the academic community, are in interplay with the teachers’ experiences, as discussed in this article. According to Wenger (2000), it is in this interplay that learning occurs. Thus, if educational researchers are truly interested in knowledge building generated from partnerships between academic and school practices, there is a need to further negotiate and (re)define some competencies that are historically defined by the academic community. The enterprise of learning about practice in practice is perhaps not only a matter of introducing teachers to the academic CoPs but of further exploring our learning of what is yet unknown at the boundaries in between research and school practice.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 The PD leader plays a formal role as a first teacher and is also a doctoral student and the first author of this paper. The second author is the supervisor of the first author.