Developing a group coaching practice targeting school leaders: an action research project with a critical friend

ABSTRACT

There are various forms of coaching in the field of education that promote professional development. This article reports on an action research project exploring school leadership coaching, demonstrating how a group coaching practice is developed when involving groups of coaches and leadership groups. The coaches, organized into two coaching teams, develop their coaching practice through reflective conversations with a critical friend, the researcher. Developing a new practice requires changing the conditions that prefigure the practice and the practice architectures or the arrangements that enable or constrain the practice. Drawing on related research on group coaching, the theory of practice architectures, and data from the two-year long action research project, the findings show how a group coaching practice involving several coaches and leaders develops through three phases. In each phase, the relationships either within or between the two groups are addressed. The findings also illustrate how collaboration between a researcher and the practitioners can promote professional learning and capacity building when new practices are being developed.

Introduction

Coaching can be an answer to the increasing demands and expectations principals and school leaders are facing as school contexts become more and more complex (Vennebo et al., 2020). For several decades, coaching has been used to develop effective teams and organizations in the business world (Brown & Grant, 2010; Clutterbuck, 2007). Brown and Grant (2010) argue that it is critical that individuals and groups have a high level of systemic awareness and an understanding of organizations and their various subgroups as dynamic and complex systems. In the field of educational leadership, dyadic coaching (Huff et al., 2013) and peer coaching (Robertson & Earl, 2014) have traditionally been employed. Recently, group coaching has emerged as an approach with strong potential for developing leadership competencies (Brandmo et al., 2019).

Despite the existing consensus that school leaders need support and that coaching can be advantageous in this respect, there is still little research on the use of group coaching in the professional training of school leaders (Brandmo et al., 2019). Consequently, few models or protocols for group coaching targeting school leaders have so far been developed (Aas, 2016). Even less research is available on coaching practices where a group of coaches are coaching a group of people having a working relationship, such as school leaders. Lofthouse et al. (2022) underlines the need for more empirical research to develop greater clarity when it comes to the complex and challenging role of the group coach (Aas & Flückiger, 2016) and for more coaching practice to be anchored in research (Grant & O’Connor, 2019). Adding knowledge to the specific area of group coaching in the field of educational leadership will be the major contribution of this study.

Action research (AR) aims to be and is a ‘practice-changing practice’ (Kemmis, 2009, p. 464) and an approach to support improving professional learning (Forssten Seiser, 2020). This study is part of an AR project and investigates how a group coaching practice directed at supporting school leadership groups develops from the experiences of the coaches, which is also a little-explored area (Forde et al., 2013). I explore how the coaches develop their sayings, doings, and relatings concerning their group coaching practice as they engage in reflective conversations with a critical friend. More precisely, the development of a group coaching practice involving coaching teams and leadership groups is framed in the following research question: How does a group coaching practice develop through an action research project with a critical friend? What are the conditions enabling and constraining such a coaching practice?

The article is structured as follows. After the introduction, I describe the empirical context of the study before I present related research on group coaching school leaders. Next, I give a short outline of the theory of practice architectures, which forms the theoretical and methodical approach applied in the current study. I then describe how the research was conducted to answer the research question and present the study’s findings. The findings will be discussed within the framework of AR, practice architectures, and related research on group coaching. Finally, I conclude and foreground a few premises that might lay the foundation for a practice in which a group of coaches is coaching a group of people having a working relationship. I will suggest addressing this practice as a group-coaching-group practice.

The AR project

In the spring of 2019, the superintendent in one of the 356 municipalities in Norway established an AR project (Carr & Kemmis, 1986). In the project, the superintendent, two coaching teams, and a researcher were assigned to work together to support school leadership groups in their endeavor to change and improve school practice. The long-term goal was to improve students’ learning outcomes. Each coaching team had three members with different backgrounds in the field of education. Two were working as principals in primary and secondary schools not subject for coaching, one was a representative from the superintendent’s office and one was a representative from the local university. While the principals joined one coaching team each, the other two members participated in both teams. In sum the coaching teams’ members represented knowledge from leading in similar school contexts, system knowledge and research knowledge.

During the first two years, the coaching teams offered ten coaching sessions each to four leadership groups and six sessions each to two leadership groups. Each leadership group that was target for the coaching, consisted of leaders from one and the same school. It was the principal and his or her middle leaders the number ranging from two to four in each school. In total 19 leaders participated. All of them had formal leadership roles in the schools in question. Their leadership experience varied a lot, both in and between the schools. As of formal leadership training, one of the 19 leaders had accomplished a master’s in educational leadership (120 credits) and six had finished a national program of leadership training (30 credits). The researcher involved was characterized as a critical friend (Henriksen & Aas, 2020; Swaffield, 2004) who observed all the coaching sessions and facilitated reflective conversations between the members of each coaching team and the researcher in the immediate aftermath of the coaching sessions.

The collaboration between the coaching teams and the leadership groups started with a school assessment of each participating school conducted by the coaching teams (Abrahamsen & Aas, 2022). When the coaching commenced, the coaching practice was inspired by a specific group coaching protocol applied in the Norwegian National Leadership Program (Aas, 2017a; Aas & Flückiger, 2016). The protocol included rules and procedures for interaction among the participants as well as time management procedures. It was tested in ten countries as part of the international project Professional Learning Through Reflection Promoted by Feedback and Coaching (PROFLEC) (Aas & Flückiger, 2016; Brandmo et al., 2019; Flückiger et al., 2017). The five steps of the protocol as applied by the coaching teams included: (1) presentation of development area chosen by the leadership group (leadership group), (2) questions for clarification (coaches), (3) reflections from the coaches related to development area, (4) response and proposals for actions from the leadership group, and (5) summary and plan for further action (leadership group). In the Norwegian National Leadership Program, one coach supported by co-coaches, those being school leaders, was coaching one leader at a time. This contrasts with the local practice, where several coaches engaged in coaching a group of leaders in a working relationship.

This project finds its foundation in the tradition of practical AR, as the practitioners aim to act more wisely and prudently, so that the outcomes and longer-term consequences of the practice will be for the best (Kemmis, 2009, p. 470). According to Carr and Kemmis,

Action research is simply a form of self-reflective enquiry undertaken by participants in social situations in order to improve the rationality and justice of their own practices, their understanding of these practices, and the situations in which the practices are carried out. (1986, p. 162)

Improving school practices and coaching practices alike has permeated the entire AR project, in which the researcher and author of this study filled the role of the coaching teams’ critical friend. Critical friendship is a form of intervention found in AR (Kember et al., 1997). In this project, the critical friend facilitated the reflection process involving the members of the coaching teams, offering them time and space for reflection on the coaching sessions and introducing theory through the questions posed. However, it was left up to the coaching teams to make the changes they considered necessary to improve their coaching practice as the sessions progressed. In this study, the coaching teams’ development of their group coaching practices is investigated.

Group coaching of school leaders

Coaching is increasingly seen as an important asset to professional development in many sectors (Gornall & Burn, 2013). It may be conducted either on an individual or group basis, according to need (Brown & Grant, 2010). Interest in group coaching has grown over the last few years. Despite the dearth of research on group coaching school leaders (Aas, 2022), studies report on benefits in different areas: developing leader identity (Aas & Vavik, 2015; Silver et al., 2009), building resilience, achieving insight into behavior, and helping to develop strategies to address the complex interpersonal relationships in schools (Forde et al., 2013).

According to Whitmore (2009), coaching is about unlocking people’s potential to maximize their own performance (Whitmore, 2009, p. 10). He further argues that it is possible to coach another person to solve a problem or learn a new skill by diligently applying learned coaching methods without basing the coaching practice on an underlying coaching philosophy. The coaching will then produce some results, but it will probably not reach its full potential (Whitmore, 2009). Whitmore (2009) describes awareness and responsibility as two coaching philosophical principles. He believes that coaching for awareness and responsibility will enhance the coaching outcome for the participants. Brown and Grant (2010) and Aas (2016) include understanding others or a sense of community as a third principle central to group coaching. Awareness deals with attention, concentration, and clarity and is emphasized because one can only respond to and change conditions one is aware of (Whitmore, 2009, p. 33). The principle also includes self-awareness, especially recognizing when and how emotions or desires affect both one’s own and others’ perceptions. The awareness-raising function of the coach is, according to Whitmore (2009), indispensable. Responsibility is emphasized because obligation and commitment increase when individuals choose to take responsibility for their own thoughts and actions. There is a big difference between being ordered to be responsible and choosing to take responsibility yourself. Increasing awareness and responsibility in a work community leads to increased performance, which in turn leads to better performance. Both principles are better generated by asking than by telling (Whitmore, 2009). The last principle, a sense of community or understanding others (Aas, 2016; Brown & Grant, 2010), is central in group coaching. The dialogue must be characterized by the participants being open to opportunities, willing to give up the right to always have the right answer themselves, and able to change their points of view through the process (Brown & Grant, 2010). The sense of community is crucial to the extent to which the participants are willing to examine their own and others’ leadership practices.

Kudliskis (2019) stated that successful coaching may be achieved through the application of models or protocols. Coaching models can help to understand the coaching intervention from a system perspective and the need for structure in the coaching interaction (Robins, 2017). Models, in practice, can shape what we do, and therefore what we achieve (Wall & Perrin, 2015). In Brown and Grants’ (2010) GROUP model, each letter represents a stage in the coaching process: Goal, Reality, Options, Understanding others and Perform. The protocol used in the Norwegian National Leadership Program (Aas, 2017b; Aas & Flückiger, 2016; Aas et al., 2020; Brandmo et al., 2019) builds on the GROUP model (Brown & Grant, 2010). Aas (2022) has conducted a review of 13 studies related to the protocol from the Norwegian National Leadership Program. Combined, the studies show that group coaching supports learning and professional development at the individual and group levels. In facilitating the reflection process, the group coach is supported by the coaching protocol. The feedback from principals having the role of co-coaches constitutes support, possible solutions to leadership challenges, and is personal, contextual, and motivating, and as such, it promotes developing leadership identity and trying out new practices.

Theories of practice architectures

The theory of practice architectures (Kemmis et al., 2014; Mahon et al., 2017) positions practice as intrinsically social and locally enacted and offers insight into the dynamic interactions between the individuals. In this study, the individuals are the coaches conducting their coaching practice in the system or arrangements where the practice is taking place. According to the theory of practice architectures (Kemmis et al., 2014), practices are formed by the complex dynamics of cultural – discursive, material – economic, and social – political preconditions. These three clusters of preconditions shape and give content to the ‘sayings’, ‘doings’, and ‘relatings’ – that is, the language, activities, and social relations to the participants involved. The group coaching practice is expressed as discourses between the coaches and their critical friend in semantic space, as activities that the coaches decide to do to change the coaching practice in physical-time space, and as particular interactions between the coaches themselves and between the coaches and the leadership groups in social space. This study explores the members of the coaching teams’ sayings, doings, and relatings to gain insights into the conditions that enable and constrain the group coaching practice. The premise of this study is that the participating coaches are best suited to develop a group coaching practice that enhances the school leaders’ ability to lead school development in their respective schools. The theory of practice architectures (Kemmis et al., 2014) is used as a methodological resource to identify and examine the coaches’ understanding of their group coaching practice (sayings), the changes they make to develop their group coaching practice (doings), and how they connect and interact with their co-coaches and their surroundings (relatings). It is also used as a theoretical resource to understand or problematize a group coaching practice involving several school leaders and coaches as it evolves.

Methods

The participants in the AR project included two coaching teams, six leadership groups, one superintendent, and one researcher. The superintendent selected the leadership groups in dialogue with the schools’ principals and established the coaching teams.

The study’s research question implies a longitudinal approach, and gathering data occurred over a period of two years. Table 1 provides an overview of the data on which the study builds.

Table 1. Overview of the data.

Data collection areas/arenas	Occasions	Data material
Reflective conversations between the coaching team and a researcher after each coaching session	4 x 10 and 2 x 6 occasions for ½ hour each	Field notes from the researcher
Written reflection notes	4 x 3 (after 6, 12, and 24 months)	Reflection notes from each of the members of the coaching teams

The reflective conversations took place following ten coaching sessions at four schools and six coaching sessions at two schools, 52 conversations in total. The conversations were framed by broad questions regarding the coaching practice asked by the critical friend. This was done to stimulate the coaches’ understanding of the coaching practice. Each coaching session lasted for three hours, while each reflective conversation was conducted in less than half an hour. The researcher’s field notes from the reflective conversations included detailed information about what was said and by whom and, as far as possible, the listeners’ reactions to what was expressed. The written reflection notes were collected from each of the four members of the coaching teams on three separate occasions, in the beginning, middle, and end of the AR project. Asking the coaches to conduct the written reflection on a regular basis had a twofold purpose: to capture the participants’ expectations of the coaching and their experience with it. To get targeted feedback, questions concerning expectations for and outcome of the coaching sessions, experiences from the coaching sessions, and from having a critical friend throughout the coaching process were raised.

The reflection-based data from the AR project that this study draws on focuses on the reflections of coaching teams’ members on their practice (sayings), the changes they make to their practice (doings), and how they relate to each other and the leadership groups (relatings). The analysis was inspired by Lofthouse (2018) and largely followed the following three steps:

Step 1: For each of the reflective conversations, key dimensions of the coaching practice were noted using the theory of practice architectures (Kemmis et al., 2014) as an organizing frame: ‘sayings’ (the semantic space) refers to the statements used by the coaches to describe their coaching practice and the context of the coaching in the recent session, ‘doings’ (the physical space) refers to how the coaching was undertaken and changes made from last session, and ‘relatings’ (the social space) refers to the nature of the roles adopted and the relationships that evolved between the coaches and between the coaches and the leadership groups.

Step 2: Using the notes describing key dimensions from Step 1, broad themes addressed in the reflective conversations across the six schools were identified to constitute three major phases in developing the group coaching practice involving several coaches and leaders.

Step 3: Examining the reflection notes for additional information about the phases identified in Step 2.

All participants in the AR project volunteered and gave written consent for the research. The Norwegian ethical guidelines for social science – based research, provided by the National Committee for Research Ethics in the Social Sciences and Humanities, were followed throughout the research. The Norwegian Centre for Research Data approved the treatment of the data in the project. To ensure anonymity, the name of the municipality, the schools, and the members of the coaching teams are not disclosed. Some limitations of the study should be mentioned. The descriptions and analyses presented here are connected to specific coaching teams. Although the findings might be examples that contribute knowledge to similar situations and contexts, more comprehensive observations of practices connected to group coaching for school development are needed and would enrich current research on the subject. The presentation includes quotes from the participants to illustrate the main findings. The quotes were translated from Norwegian to English, and, some meaning may be lost in the translation process. To increase the quality of the work, I have used member checking (Lincoln & Guba, 1985), which means that the coaches in this case have read or been presented the empirical findings to ensure both accuracy and ethical treatment. I further argue that AR, with its emphasis on intervention, is ethically valuable because it seeks to give something back to the participants that can guide them in their further practice.

Findings

I will now turn to a more specific clarification of what the members of the coaching teams addressed in the reflective conversations and in the reflection notes as they developed their group coaching practice. The themes that emerged tended to recur through the AR project, but in the presentation, each theme is addressed in the phase of the AR project where it was most prominent. The phases are as follows:

Phase 1 (months 1–6): Structuring the interaction between the coaches

Phase 2 (months 7–12): Structuring the dialog between the coaches and the leadership groups

Phase 3 (months 13–24): Developing the dialogue through inquiry

In the account of each phase, there is: (1) a description of what the coaches highlighted the most in the reflective conversations, (2) a description of the coaches’ attention to the same theme in the adjacent reflection notes, (3) a description of the challenges the coaches addressed (retrieved from both reflective conversations and reflection notes), and (4) a description of the coaches’ opinions on the presence of a critical friend. Each presentation ends with a summary of the main development in the coaching practice in that phase of the AR project.

In the presentation of the empirical findings, I refer to the different reflective conversations and reflection notes through coding. The four coaches will be distinguished from one another by the codes C1, C2, C3, and C4. The reflective conversations are marked with RC, the number of the RC at each school is listed from 1–10 and the respective school ranged from School A to School F. An example might be RC5-A-C1, which gives the following information: Reflective Conversation number 5 at School A with Coach 1. The reflection notes will be marked with RN, and the number of the reflection notes is listed from 1–3. An example of this coding could be RN2-C3, representing Reflection Note number 2, written by Coach 3.

Phase 1 (months 1–6): structuring the interaction between the coaches

Phase 1 constitutes the first six months of the AR project. During this period, the reflective conversations commenced with the critical friend posing the question, ‘How did you experience today’s conducted coaching session?’ The analysis showed that the initial focus was directed at how to work in a coaching relationship. This was an issue in 10 reflective conversations across the schools in the first phase.

It became evident early on that defining roles within the coaching teams was necessary (Abrahamsen & Aas, 2022). Opening the reflective conversation in the second coaching session, the coaches expressed that the session had been unstructured and messy (RC2-A-C4) and that they had to stop interrupting each other (RC2-A-C1). Statements such as, ‘It is a good idea to come to an agreement about who is doing what’ (RC3-A-C2, RC3-B-C1, RC3-C-C3) occurred repeatedly. This was also addressed in the first reflection notes, where one coach wrote: ‘We need to be more coordinated to become better coaches’ (RN1-C2).

Another coach expressed:

We have agreed that we must define our different roles in the coaching teams and develop structures that can guide and strengthen our work for the better. We are working on it, but there is a lot to do in a hectic everyday life, and we should probably look even more into this. (RN1-C1)

Despite the fact that the coaches decided on and openly declared a division of roles in Phase 1, wherein each of them carried the main responsibility for specific tasks in the coaching session, being several coaches still was challenging. In the reflective conversations throughout the AR project, the coaches expressed that they found it hard to keep an overview of each other’s contributions (RC7-B-C2, RC7-A-C4), and that other coaches’ intervention in their specific tasks made them lose their train of thought (RC4-B-C3, RC7-B-C3). This is also described in the first reflection note. One coach wrote: ‘I have experienced that planning and conducting coaching sessions when three coaches are involved is challenging. We must plan the sessions in more detail, and we must have more defined roles’ (RN1-C3). Another coach described that ‘there is a lack of time set aside for this work’ (RN1-C2).

Unsolicited, the role of the critical friend in developing the coaching practice was addressed in the first reflection note:

It has been very helpful to collaborate with a researcher as a critical friend. We have had to evaluate ourselves during the process. In a busy everyday life, this would probably not have been done, and even if we did it on our own, it would probably not have become as important and valued. (RN1-C1)

The reflections on the coaching practice in Phase 1 of the AR project led to defining the role of each coach and structuring the interaction and working relations within the coaching teams. Coordination between three coaches took a lot of planning and preparation, and lack of time was an issue. The support from a critical friend seemed to contribute to the coaches’ reflections on their practice (sayings), to changes to their practice (doings), and to how they interact with each other in their coaching practice (relatings).

Phase 2 (months 7–12): structuring the dialog between the coaches and the leadership groups

In Phase 2 of the AR project, 10 of the reflective conversations across the schools were dedicated to developing a local group coaching protocol, using the protocol from the Norwegian National Leadership Program (Aas, 2016; Aas & Flückiger, 2016; Brandmo et al., 2019) as a starting point. The local protocol had three steps: Step 1. Opening (parameters for the session, long- and short-term goals, conducted in-between actions [that is, new practices tried out since last session], and coaching issue). Step 2. Theory and reflection on coaching issue, and Step 3. Closing (new in-between actions and evaluation of session). Throughout Phase 2 of the AR project, each of the steps was deliberated on and clarified.

In this phase, the critical friend continued to pose questions about how the coaches experienced the conducted coaching session. Satisfaction with the new protocol was expressed (RC3-D-C1); however, the coaches also discussed the time allocated to the different steps (RC4-C-C4, RC6-B-C2) and the need to facilitate for meta-communication about the progress of the coaching session when many people were involved (RC3–4-C1). The coaching team started to display the protocol on a digital screen, demonstrating to everyone involved how the session was progressing and using it to get back on track if necessary. This was a response to what one coach expressed in the second reflection note: ‘We must explain what the leadership groups can expect, so that they are better prepared for what is happening in the coaching sessions’ (RN2-C4). Allocating time for two breaks to conduct separate talks in the coaching team and the leadership group became important for calibration and moving forward through the different steps in the protocol (RC3–4-C2). In a reflective conversation, one coach described that the breaks were needed for the coaches to consult one another (RC4–3-C1). An excerpt from the second reflection note describes how the new protocol structured and improved the dialogue:

First, I would like to highlight the good conversations taking place. I have seen a development in the way we have been able to dialogue with the leadership groups we have met with throughout the year. We have come up with a coaching protocol to facilitate the collaboration between the coaching teams and the leadership groups. (RN2-C3)

A challenge that the coaches faced was finding enough time to discuss their coaching practice and address necessary changes: ‘It is a shame that it always gets so hectic that we are unable to reflect sufficiently on our critical friend’s input after each coaching session. I sometimes experience that we do not take the input enough into account in the next coaching session’. (RN2-C4)

When the coaches were asked to address their experience with their critical friend in Reflection Note 2, one coach wrote:

As coaches, I don’t think we fully realized the extent of the work we were about to do. We did not have a mutual understanding of what we were doing, which coaching protocol we should apply, what its content should be, which roles each of us should take on, and so on. Our critical friend asked questions for reflections that helped give direction when we at times went astray and lost sight of the goal. Without this, I’m afraid that we wouldn’t have been able to see the potential for improvement so quickly, and in some cases, this could have led to us doing more harm than good. (RN2-C2)

The major development in the group coaching practice in the second phase of the AR project was structuring the dialogue between the coaches and the leadership groups through a group coaching protocol displayed to everyone involved throughout the coaching session. Phase 2 was in such dedicated to structuring the interactions and relations between all the involved participants.

Phase 3 (months 13–24): developing a dialogue based on the needs of the leadership group

While the previous phase was about structuring the dialogue, Phase 3 was dedicated to inquiry to make sure that the content of the dialogue was meeting the needs of the leadership groups in their school context. In this phase, the critical friend questioned how each coaching session contributed to fostering awareness, responsibility, and a sense of community in the leadership group. This question led to responses in 20 reflective conversations across the schools on the leadership groups’ ownership of the coaching issue and the in-between actions, the latter consisting of the new practices that the leadership groups were obliged to try out between the coaching sessions.

From time to time, the coaches reflected on why the leadership groups sometimes came to the coaching sessions without having conducted the in-between actions (RC6–1, RC8–3). At other times, they wondered whether it was the entire leadership group or the principal who had suggested the coaching issue or conducted in-between actions (RC4–5-CO2, RC9–4-C2). During Phase 3, the coaching teams adjusted their practice to allocate more time to explore the coaching issues, the in-between actions, and concepts suggested by the leadership groups. At the same time, they reflected on how this seemed to raise mutual awareness and equal responsibility in the leadership group (RC3–5-C2).

In the reflective conversations, balancing support and challenge in group coaching also became an issue. One way to offer support was to make sure that the leadership groups received coaching on an issue they had suggested and in such make the coaching session relevant to them (RC7–2-C2). The coaches’ experienced that the leadership groups took ownership and made their own choices when the coaching team facilitated them to do so (RC3–6-C1) and when they made sure to address the leaders’ school context (RF10–2-C2). They also had to challenge the leaders to be more concrete (RC9–3-C1), but found it trying to challenge without being interpreted as critical toward the leadership group. This was addressed on several occasions throughout the coaching period (RC8–2-C2, RF8–2-C1). One coach wrote:

I hope and believe that I have become more aware of how I ask questions, aware of asking more open-ended questions. This so that leaders, to a greater extent, own and formulate their needs when it comes to the content of the coaching session in cooperation with us, as well as the proposals for in-between actions. (RN3-C2)

On having a critical friend, one coach wrote in the last reflection note:

It has been useful to have someone observing our practice from the sideline and giving us relevant input and questions for reflection after every coaching session. When it comes to how we interact and distribute tasks within the team, how we contribute to the ongoing coaching, and how we appear to the leadership groups. The reflections have helped us keep direction and explore what the leadership groups want to focus on, to contribute to the leadership groups’ ownership of the new practices they are to try out before the next coaching session and to find their own solutions to the greatest extent possible. (RN3-C2)

The reflections on the coaching practice in Phase 3 of the AR project led to a clarification of how inquiry balancing support and challenge could contribute to coaching sessions meeting the needs of the leadership groups. In this phase, the focus was on the interactions or relations within the leadership groups situated in their school context.

Summary of the findings

The analysis of the reflective conversations and the reflection notes made it clear that the coaches developed their sayings, doings, and relatings (Kemmis et al., 2014) concerning their group coaching practice throughout the entire action research project (Table 2). Nevertheless, the aspect of social space in the practice architecture can be seen as generating most discussion: first directed at relatings within the coaching team, next relatings between the coaches and the leadership groups, and, finally, relatings within the leadership groups. The reflection notes showed that the members of the coaching teams found the support from their critical friend decisive because both time and space were allocated to reflection on practice, which enabled them to a) evaluate their own practice, b) develop a mutual understanding of their coaching practice, and c) develop and try out new practices.

Table 2. Identified changes in the coaching practice.

Group coaching practice	Sayings	Doings	Relatings
Phase 1	Structuring the interaction	Division of roles in the coaching team	Within the coaching team
Phase 2	Structuring the dialogue	Group coaching protocol	Between the coaching team and the leadership group
Phase 3	Coaching to meet the needs of the leadership group	Inquiry to support and challenge the leadership group	Within the leadership group in the context of their school

Discussion

Developing a group coaching practice involving several coaches and leaders was accomplished through three different phases in this AR project. The analysis showed that the coaches first decided on a division of roles between themselves, still finding it challenging to take advantage of actually being three coaches throughout the AR project. Next, they developed a group coaching protocol in three steps to facilitate the dialogue between themselves and the leadership groups, still not being able to find enough time to reflect on and improve their practice. Finally, they directed their attention toward the leadership groups and their needs, still finding it demanding to balance support and challenge in their inquiry.

In this section, I will discuss the findings connected to developing a group coaching practice involving several coaches and several leaders against theory on the GROUP coaching model (Brown & Grant, 2010), the protocol from the Norwegian National Leadership Program (Aas & Flückiger, 2016), and the coaching philosophical principles (Aas, 2016; Brown & Grant, 2010; Whitmore, 2009). These theories have so far been applied to coaching practices involving one coach who is coaching one leader with support from co-coaches. I will especially investigate enabling and constraining factors in a practice where groups are coaching groups, in contrast to a coaching practice involving fewer people.

Phase 1: ‘we need to be more coordinated to become better coaches’

In group coaching, there is a need to establish ground rules from the outset that are shared by all the group members (Aas & Flückiger, 2016). The analysis showed that when several coaches are involved, they must define their roles internally between themselves and externally toward the leadership groups before they can work on the content of the coaching.

For the coaching to be productive, the coaches’ competencies were found to be particularly important (Aas, 2016; Bloom et al., 2003). Knowing what it is like to be a school leader and having knowledge of leadership challenges within education seems to be crucial in coaching school leaders (Aas, 2017a). Assembling a group of coaches, it is possible to combine coaches with a nuanced understanding of educational contexts. In this way, the coaches can have the advantage of research knowledge, system knowledge, and knowledge from leading similar school contexts. In coaching situations, group coaches can share their experiences, knowledge, and skills reciprocally across professional boundaries. Defining roles is part of establishing internal collaborative professionalism. Using the coaches’ education and professional work experiences might be a starting point when roles are defined and attributed. Such prior experience might guide how responsibilities are shared. If several coaching sessions with the same group of coaches are to take place, it might be necessary for one of the coaches to coordinate practical matters in between the sessions, for instance, to convene meetings among the coaches to plan coaching sessions.

There is a lot of focus on building trust (Whitmore, 2009) in the coaching situation. It is also critical that trust is established among a group of coaches. Trust can be generated over time and thus provide an environment for professional learning and change. Members of a group of coaches must listen carefully to one another, as well as to the group they are coaching, and consider how to take advantage of the knowledge and personal competencies of each coach.

Defining roles seems to be an enabling factor in a practice in which a group is coaching a group. Facilitation can be shared between the coaches and, as such, expand each coach’s capacity to prepare questions for reflection. However, coaches must constantly consider when to intervene in the ongoing reflections and when not to. If they disrupt their co-coaches, they could be obstructing instead of fostering progress in the coaching session. The coaches need to be coordinated, which assumes that time is set aside to prepare and plan the coaching sessions, something that might be a constraining factor in a hectic everyday life.

Phase 2: ‘a protocol should facilitate collaboration’

Applying protocols or models is one way to ensure that coaching progresses in the right direction and that the coaches include all stages of the coaching process (Kudliskis, 2019; Robins, 2017). The protocol underpinning the structure in the coaching sessions in the AR project came from the Norwegian National Leadership Program (Aas, 2016; Aas & Flückiger, 2016). Aas (2016) suggest that local and individual adaptations should be conducted when existing protocols or models are applied in new contexts. Adaptations can be necessary, as in this case, where a) a group of coaches are coaching a group of school leaders, b) the coaching is implemented to support school leaders in leading school development, and c) there is a long-term goal for the entire coaching process, as well as a short-term goal for each session. The analysis showed that the group coaches in the AR project adapted the protocol from the Norwegian National Leadership Program to fit their needs.

Experienced and inexperienced coaches use coaching models or protocols differently (Robins, 2017). It appears that a group of coaches, inexperienced with coaching together, need a more detailed protocol adapted to their different roles and the long-term coaching goal to scaffold their coaching. In their situation, the protocol outlines the interaction between the group coaches and the leadership group and includes structure, content, timeframe, and division of roles. Displaying the protocol on a screen for everyone to see throughout the coaching sessions is one way to facilitate predictability when several participants are involved. It can lay the foundation to engage and be complicit in the process. Displaying the protocol also facilitates moving between the different steps, when and if necessary. As such, it secures awareness and complicity through the coaching process.

Developing a context-sensitive protocol seems to be an enabling factor in a practice where a group is coaching a group. This is possible when several coaches are involved, and time is set aside to reflect on what coaches and leaders need in their context. Lofthouse (2018) found that when a coaching relationship is given time to develop and the coaches’ understandings are nuanced, the practice of coaching can be more than an instrumental process. Instead coaching can create opportunities for co-construction (Lofthouse et al., 2010). As the coaches get more experienced coaching together, it is imperative that a protocol does not become prescriptive or rigid. The extent to which it meets the needs of those receiving coaching needs to be continuously assessed. If not, a protocol might become a constraining factor.

Phase 3: ‘what do the leadership groups want to focus on?’

In a group coaching practice involving several coaches and leaders, the practice needs to facilitate the integration of multiple voices. A coaching practice of collaborative professional inquiry characterized by a collaboratively identified problem (coaching issue) and a subsequent activity (in-between actions) can calibrate colleagues in a leadership group and foster awareness, responsibility, and sense of community (Aas, 2016; Brown & Grant, 2010). When coaching issues are suggested and in-between actions are described by the leadership group prior to a coaching session, as in this AR project, the integration of multiple voices in the coaching sessions becomes even more crucial. Inquiry into each leader’s understanding of their school practice and context can encourage every member of the group to contribute to the dialogue, understand the issue at hand, and be jointly responsible for improving the current practice. Such a group coaching practice is a way to avoid one leader formulating the coaching issue on behalf of the others, an uncertainty that exists when a group is being coached. This makes it even more important to facilitate joint deliberation to increase collective willingness to carry out changes.

It is easy to assume that coaching is supporting someone in their process of learning, forgetting that it is also challenging on individual and group levels. A balance can keep leaders motivated to try out new practices. Group coaching school leaders is about supporting and expecting transformation. The group coaches have to make use of both support and challenging skills, making sure that the inquiry identifies the leaders’ capabilities to improve their own practice, and at the same time making sure that there is no critique in the inquiry. It seems like it is easier to support and challenge the leadership group when there is an agreement on and understanding of the coaching issue and when it is recognizable and relevant to every member of the leadership group. Allowing enough time to inquire into long- and short-term goals, the latter including each session’s coaching issue framed by the leadership group, seems to be an enabling factor in a practice where a group is coaching a group, as is raising awareness, responsibility, and a sense of community in the leadership group when it comes to the practices that are to be tried out. If the inquiry is conceived as critique by the group of leaders being coached, it might be a constraining factor destroying the balance between support and challenge in the coaching session.

Conclusion

The purpose of this study was to examine how a group coaching practice involving coaches and several school leaders in a working relationship developed through an AR project with a critical friend and, at the same time, study the conditions enabling and constraining such a coaching practice. I would like to suggest addressing this practice as a group-coaching-group practice, for lack of a better term, and because it is necessary to distinguish a practice involving several coaches and people being coached from other types of practices.

The practice architecture of a group-coaching-group practice consists of regulating the participants involved in a certain manner (doings), here through a protocol developed for a group-coaching-group practice, from certain roles and in relation to each other (relatings), and with language through reflection (sayings). In developing a group-coaching-group practice, the importance of material-economic preconditions comes to the fore, especially having enough time to conduct a practice involving several coaches. At the same time, it seems that reflections with a critical friend are an enabling factor in the cultural-discursive dimension. The coaches’ motivation to develop their practice can be found in the social-political dimension of the practice architectures (Kemmis et al., 2014) and includes the relatings between the participants involved: first the coaches, then the coaches and the leadership groups, and, finally, in the leadership group situated in their specific school context.

Developing a leader coaching practice in which groups are coaching groups, it seems necessary to travel through some phases. The group-coaching-group practice should be adapted over time to meet the needs of the coaches and the leadership groups. It also seems that the practice development must be deeply contextual and reflection-based for coaches to see the relevance and commit to developing their practice together.

The implications of the study can be summed up in the following premises that might guide the coaches in their group-coaching-group practice: (1) Reflect on how to work in a coaching relationship and clarify a division of roles among the coaches. (2) Decide on a group coaching protocol that can structure the dialogue, especially when the participants are new to one another. (3) Facilitate the integration of multiple voices in the leadership group through inquiry that both supports and challenges to create an awareness of the school context that fosters responsibility in the leadership group to try out new practices (in-between actions). (4) Allow the group-coaching-group practice to evolve over time. (5) Support professional learning and development through AR and critical friendship, if possible. Reflection on coaching practice can be an important element for coaches to develop both individually and in groups.

Further research should explore the extent to which these premises can be successfully adopted by other coaches involved in a group-coaching-group practice and investigate the views of groups being coached by a group of coaches, as this research only offers the perspectives of the coaches.

Disclosure statement

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

Data availability statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to information that could compromise the privacy of research participants.

Additional information

Funding

The work was supported by the Norges Forskningsråd.

Notes on contributors

Lena Abrahamsen

Lena Abrahamsen is a doctoral candidate in Educational Leadership in the Department of Education at UiT The Arctic University of Norway. She is an experienced teacher and principal and has since 2019 worked on her public sector PhD project. Her research interests are in the areas of leadership and school development, professional development of schoool leaders and group coaching.