Video-Based Action Learning and Research: Increased Transformative Capacity among Team Leaders of a Youth Care Protection Agency

Public organizations need to learn and evolve continuously to keep up with emerging complexities. This may require a transformational organizational change, including culture, strategy, structure and working methods for service delivery. The aim of the study is to understand how a video-based reflection method could support an action learning (AL) and action research (AR) process to enhance transformative capacity among a group of team leaders in a changing youth care organization. Sixteen team leaders participated in the video-reflection process. The steps were: (1) all leaders were filmed leading a team meeting; (2) each leader reflected on a peer’s video; and (3) reflections were analyzed, and themes were abstracted. (4) Related to these themes, video fragments were compiled to share with the leaders in Focus Group Discussions (FGDs), where the leaders discussed the videos and formulated key lessons. (5) Key lessons were summarized in a report. Data collection included: video-recordings, evaluation sheets and field notes acquired during the process of video-reflection and seven months later, interviews with team leaders (n = 11) about their learning experiences. This study shows greater transformative capacity in most of the team leaders. Awareness of their work practice, as well as the process of reflexive monitoring during the FGDs, contributed to widening the scope of their reflective reality, and thus their sense of agency in adapting their practices, such as moderating their occasionally controlling behavior, being able to structure the team meetings effectively, and understanding when to use specific leadership roles.


Introduction
In the field of public welfare, organizations are increasingly pushed to learn and evolve their organizational practices in order to keep up with the emerging complexities of contemporary society (MacArthur et al., 2011;Packard et al., 2017). They might do so by gradually transcending a particular organizational culture, or by initiating radical organizational transformation (Rauws, 2017). Organizational transformation, however, commonly poses important challenges for leaders who need to make sense of a new direction, in the face of a high level of unpredictability (Lüscher & Lewis, 2008). Organizational transition theory emphasizes that in situations of complex change, leadership should not so much focus on adapting organizational practices, but on shaping the conditions for co-evolution and a context in which organizational practices can be continuously adapted (Grin et al., 2018). It means that leaders should empower their staff to collectively learn and implement new solutions (Mendes et al., 2016), thus enhancing their 'transformative capacity'.
Transformative capacity is a central concept in the Structuration Theory developed by Giddens (1984). It provides a useful lens to organizational transformation, as it acknowledges the context of rules and norms that help to reproduce social practices, as well as the agency of actors to contest and transform these practices (Canary & Tarin, 2017). Within this theory, the capacity of professional staff to 'act otherwise', concerns the way actors mobilize resources to realize learning and change within their organization (den Hond et al., 2012).
In relation to the above, Argyris (1993) suggests that leaders might benefit from tools that enable reflection in practice, such as collaborative 'action research' tools, to help professionals learn. A combination of Action Learning (AL) and Action Research (AR) can help leaders find new practices in settings of transformational change (Coghlan, 2019;Coghlan & Coughlan, 2008). ALAR, the acronym used by (Coghlan & Coughlan, 2008), is focused on studying and transforming practices in action, and includes principles of reflexivity and co-created knowledge production or Mode-2 research. In AL, the focus is on learning from and through action (Coghlan, 2019), such as developing a new (leadership) practice with the help of action and reflection cycles. In AR, a systematic research approach is used to develop grounded research activities that in propel change and further investigation. Both research and learning are required to reflexively examine and transform organizational work, and studies using ALAR have involved managers and their capacity to learn and transform in organizational settings (e.g. Shani & Coghlan, 2019;Wyton & Payne, 2014;Yström et al., 2019).
The current study is specifically focused on enhancing leaders' transformative capacity, by evaluating the use of a peer video-based reflection approach. Video-based reflection methods situate research and learning in the real world (Vince & Earnshaw, 2012) and offer opportunities to analyze the rich and detailed material of work practices (Heath et al., 2007). By showing fragments of real-time actions, and how these are organized in their work environment, video-based reflection encourages people to learn both from their own and others' practices, thus potentially enhancing their capacity to transform. Video-based reflection is in line with both ALAR and ST, as it helps professionals reflect on dynamic data from their own field of work, creating deeper understanding and more meaningful feedback loops.
Few studies to date have analyzed how the tool could contribute to a process of learning and research, which ultimately also leads to transformed leadership practices. The aim of this study is, therefore to understand how a video-based reflection method could support an ALAR process to enhance the transformative capacity of team leaders in a changing youth care organization.

Transformative capacity
The concept of the transformative capacity of human agency that enables change takes a central place in Giddens' Structuration Theory (ST) (Den Hond et al., 2012), which contends that social practice involves both structure and action, and that social actors have different sources of power that help them navigate and engage in social practices, by producing, reproducing and transforming them, as part of a group or organization. It is often used in studies of organizational change, because it offers the theoretical framework through which to examine the complex, dynamic relationship between human agency and social institutions, such as, in the case of this study, a youth care protection organization in Amsterdam (Canary et al. 2017).
In many ways ST explores how people learn through monitoring their actions reflexively. It does so by focusing on the way agents, while bounded in certain structures, draw on their knowledge to potentially change that structural context. Knowledgeability, according to Giddens, implies a sense of reflexivity and mindfulness, and is instrumental in the production and reproduction of practices. It is defined as: 'Everything which actors know (believe) about the circumstances of their action and that of others, …. including tacit as well as discursively available knowledge' (Giddens, 1984, p.375). Knowledgeability refers to the actors' capacity to understand how their actions are motivated and what consequences they might have, which enhances their sense of 'agency'. Giddens suggested that knowledgeability includes three dimensions: discursive consciousness, practical consciousness and unconscious motives. Discursive consciousness refers to the level of knowledgeability in which humans can express and discuss their thoughts, emotions and reasons for actions (Thuo, 2013). It includes the elements of their conscious (social) reality that can be translated and communicated and can thus also be treated as explicit knowledge. At this stage of discursiveness, actors can reflect on why they do or do not do certain things. It is through discursive consciousness that actors can make themselves understood, and collective knowledge is constructed in time. Thus, discursive consciousness and collective knowledge hold social practices together (Wheeler-Brooks, 2009). The second level of knowledgeability, practical consciousness is related to practical knowledge, or knowledge that is not verbalized or expressed. Associated with practical consciousness are 'routines', since with time, actions as well as interactions between actors can become part of the environmental structure (bearing in mind that structure is not external to human action, but is what shapes and gives form to social life), and result in routinization: 'the habitual, taken for granted character of the vast bulk of activities of day-to-day social life' (Giddens, 1984, p.376). Practical knowledge can be made discursive if humans explain their actions and the reasons for them. However, the third level of unconscious motives, related to tacit knowledge, is inaccessible and usually remains hidden from agents.
Through the process of ALAR, professionals can be engaged in monitoring their actions and routines, thus enhancing their knowledgeability and transformative capacity. In this study, we aim to understand how video-based reflection can be used to support this process in practice.

Video-based reflection to enhance transformative capacity
While videos have been extensively used in student teaching, particularly in the fields of medicine (Darbyshire & Baker, 2012), nursing (Oh et al., 2012) (Cappiello & Vroman, 2011;Oh et al., 2012), education (Decoster & Vansieleghem, 2014;Membrives et al., 2016;Rondina & Angelucci, 2017), counseling (Gary & Grady, 2015;Higgins & Dermer, 2001) and various other disciplines (e.g. Hudock Jr & Warden, 2001;Kuriansky et al., 2010), the focus of videobased reflection lies more specifically in creating an interactive and dynamic learning process. Video as a tool is here primarily used to help individuals and organizations transform in order to learn how to improve their practices. It relates to the concept of ALAR as it involves a group-oriented form of developing competencies and organizational culture, while reflecting on and studying people's actions.
The visual representation offered by video-based reflection, including people's gestures, speech, movements, coordination, use of objects, routines, collaboration, roles, and the highly contingent nature of these phenomena in their local context, is expected to help people see and discuss how practices are produced and reproduced in action. Rather than replicating existing practices, video-based reflection is used as an instrument to transform these actions both collectively and individually (Puentedura, 2013). The video material is discussed and analyzed with reference to the ambitions of radical innovation and change that drive an organization. In other words, video-based learning is used as a tool to leverage the potential of organizational change as a whole towards a particular novel vision. Video reflection as an intervention is thus used to activate ALAR, which depends on four activities, including (1) observation (what is really happening?); (2) analysis (collectively drawing attention to the dynamics of the local environment and how things happen); (3) reflection (deeper understanding of values and norms that motivate certain actions); and (4) adjustment (naturally responding differently). Video-based learning is particularly valuable in making phenomena tangible so that the material can be analyzed, reflected upon, and adjusted more effectively.
While the approach of combining video with some form of guided reflection has been used before, particularly in the context of teacher training (e.g. Shanahan & Tochelli, 2014;Tripp & Rich, 2012), the uniqueness of this study lies in seeking to support radical organizational change through improved leadership practices. Those who are involved in such a process are consistently learning about what they are doing, as they are doing it, and try to adapt accordingly. This happens partly through the production of, and navigation between, semiotic frames (see Lussi Borer & Muller, 2016). Actors involved in a learning process relate to each other through the video artifacts by reflecting on (1) the meaning of the viewed activity, and (2) the relationship of the viewed activity to another, often their own, activity. The actors involved view a peer's filmed activity, and thus a process is set in motion in which semiotic frames are produced that relate to the activity being viewed, which then creates the potential for transformation.

Methodology
This study involved an ALAR process supported by a video-based learning tool. In this section we describe the intervention with the video-based reflection instrument, and how it reflects both action learning and research aspects.

Study context
This study was undertaken at the Youth Care Protection Agency in Amsterdam in the Netherlands. The organization went through a major crisis in 2008: it had financial deficits, was placed under supervision by the health authorities and had a low client satisfaction rating (Dinkgreve, 2017). This crisis precipitated a need to fundamentally and radically change the organization (from an organization based on New Public Management (NPM) to becoming a learning organization) to improve children's protection and prepare for the decentralization from the provincial government to the municipalities in 2015. The primary mission or purpose of the organization was formulated as 'every child safe'. An essential element here is that the employees were asked to be more reflective in their work and work together in teams. Similarly, leaders needed to learn how to guide their staff to these ends. (Van Veelen et al., 2018).
In this new approach, called Intensive Family Case Management (IFCM), 1 team leaders are responsible for guiding a learning process among their staff by facilitating weekly meetings with a team of six to seven family managers, a psychologist and a senior supervisor. In these meetings, team members are expected to reflect and decide on cases together, most of which are complex and challenging because they concern children's safety and the privacy of family life (Busschers et al., 2015;Dinkgreve, 2017). Of essential importance is that leaders stimulate the continuous reflection and learning process among team members, especially as the guiding principles of the IFCM are continuously redefined in the dynamic practice of youth protection work. This requires a very different type of leadership practice, in which leaders steer and support the team, rather than guide them with traditional, 'topdown' management strategies. It is within this context that this study, and the tool of videobased reflection, was introduced.
The video-based reflection method comprised five distinct phases (see Fig. 1).
1. Team meetings (n = 16), lasting approximately two to three hours, were conducted and recorded (action). During the recording, the focus was on the leader as s/he was guiding the meeting in which the team members discuss their actual cases (families). 2. The videos were shared with four groups of four peer team leaders, who were then asked to evaluate a video of their respective peer (reflection). Reviewers were asked to indicate (1) eye-openers ("I see something interesting and helpful"); (2) remarks ("I have something to say or add to this"); and (3) questions ("I wonder why you are doing this?"). 3. Both the videos, as well as the peer's evaluations, were analyzed by the principal researchers to derive common themes (research). The video clips connected to these themes were extracted and edited to form one new video for each group of team leaders (n = 4).
4. These compressed videos were shown and discussed in-depth in group discussions with the leaders in their respective groups (reflection and learning) and one or two members of the board. First the eyeopeners were shown, which showed that a lot was going well, and helped to create a safer environment. The other themes were discussed by first talking about the challenges, followed by the underlying motives and then by strategies to overcome these challenges. 5. The results of the discussion groups were summarized in a report written by the principal researcher and shared with all the team leaders for validation.
The entire process lasted for about five months.

Data collection
During the action research process, data sources included the video-recordings, evaluations, field notes and interviews. Fifteen team leaders, four men and 11 women (average age 40) took part in this study. All of them were graduates, and some also had a postgraduate degree. Most already had several years' experience in guiding teams before participating in the study, and had guided teams for about seven months with the new approach to working as well as having participated in the video-reflection method.
Semi-structured interviews were conducted to explore the impact of the video-based reflection method as well as assess the knowledgeability and discursive awareness (Gray, 2013). Team leaders were invited by email and eleven (out of the 16) were able to participate. The non-participants had left the organization (n = 3) or cancelled the appointment (n = 2) and rescheduling was not possible in the timeframe of the research project. No reasons for cancellation were given or sought. Of the eleven interviewees, ten had watched their own video, while one said she felt too uncomfortable with watching herself. They were asked to share the challenges they were facing in their leadership role, and the changes in their approach to guiding their team. Interviews lasted about an hour. They were audio-recorded and transcribed verbatim. To reduce the principal researcher's potential interpretation bias, a member check was done by sending a summary of the interview to the interviewee asking them to validate the summary and/or send adjustments.
The first author was a policy advisor for the youth care organization and observed and participated in the organizational change process. The videos of team meetings and FGDs were reviewed and evaluations read, in order to better understand the context of what had been said in the interviews.

Data analysis
Theoretical thematic coding was used to analyze the interview data (Gray, 2013). For the analysis, all transcripts were initially read by the first two authors to gain a preliminary impression of the data. Subsequently, the interview transcripts were analyzed by both researchers using Atlas.ti software for qualitative data analysis. A combination of deductive and inductive coding was used. A coding scheme, based on the concepts of knowledgeability (Giddens, 1984), was created (see Table 1), while also allowing for any unexpected themes to come up (Fereday & Muir-Cochrane, 2006). As a third step, the first two authors reflected on the emerging codes and jointly derived a set of themes and sub-categories. These themes were discussed, validated and enriched by the data that emerged from the videos and evaluation sheets and complemented the transcripts. Finally, the entire process was checked by the last three authors to reduce potential interpretation bias and increase the consistency of the analysis. Any inconsistencies were discussed and, after reaching consensus, corrected before writing up the results (Gray, 2013).

Ethical issues
No formal ethical approval was required according to Dutch law and the researchers followed the Dutch code of conduct for scientific practice.
Privacy and a safe environment were the two most important ethical aspects of our study. The issues discussed concerned families and children, which are highly sensitive and private. Therefore, privacy protocols were used and it was ensured that names of children and family members could not be heard in the team meetings' video-recordings.
Most people are intimidated by the idea of being video-recorded. Therefore, we used a small camera to avoid creating a distraction. Furthermore, the facilitator of the videorecordings made sure that the recordings were only used internally and that team leaders Table 1 Operationalization of knowledgeability Dimensions of knowledgeability Operationalization in the context of this study

Unconscious motives
The team leader discusses and seeks to discover unknown underlying patterns and motives of her of his behavior

Practical knowledge
The team leader evaluates her/his own or other routines and behavior and can implement new or adjust routines and behavior Discursive consciousness The team leaders are discussing together and formulating meaning by giving words to routines and behavior in the video. In addition, they are discussing their underlying fears and motives in order to be able to transform these had the right to say who was allowed to see their video. We developed a protocol to ensure the team leaders could control this process rather than being controlled by it, and that they felt they were in a safe learning environment. The video-recordings were shared using a secure business network storage, to which access was managed by using the organization's active directory (Derry et al., 2010).
Prior to the interview, we informed the interviewees about the purpose of the interview and that they could withdraw at any time. Their informed consent was then requested and recorded on tape.

Results
From the interviews, it became clear that the participants had found the video-reflection method both confrontational but also enriching. The leaders appreciated the rich information they received from the videos, which is hard to achieve through written text or verbal feedback. They all felt it accelerated their learning process: 'I found it quite overwhelming; you can say a thousand words to each other after a meeting. But if you see yourself on video, the impact is much larger.' (Team Manager (TM)-4) 'The video had a real impact, I thought: wow, I really have to change some things.' (TM-10) 'You start to think how you do things because you are really visually confronted. The video hides nothing. So, it starts a change process.' (TM-8) The following section presents the findings in three parts. The first part describes the leaders' greater practical consciousness, and how aspects of the video-based reflection method helped to improve their individual practices. The second part addresses the discursive consciousness as developed in the process, and how discussion based on video material helped to foster a common language for leaders. Both parts also describe the role of unconscious motives. The third part reflects the way these three components of Giddens' theory on structuration come together in building transformative capacity among leaders in the organization. A summary of the findings in terms of leadership outcomes and conducive aspects of the video-based reflection method is presented in Table 2.

Practical consciousness
A year after the experiment had finished, all participants were asked to share what challenges they were facing in their leadership role, and also what had changed in their approach to guiding their team. The analysis of their answers showed one important, and relatively uniform, change that occurred in the leadership practice of almost all participants, largely reflecting more ability to empower their team and let go of excessive power. What did this change look like exactly?
First, participants expressed the view that their previous leadership style was focused on keeping a tight grip on time, content, and process outcomes. They had felt that this was needed in order to realize learning and quality results by controlling the content and to achieve the operational targets by controlling time and process. By seeing other ways in which their peers in dealt with these tensions, they understood that various unconscious motives also played an important role in their less than empowering leadership style. Some of these motives were formulated as either ensuring that, as a leader, you do enough to validate your role, or that you steer the team towards the (only) desirable and valid outcomes. The interviews seemed to confirm that, following the video method, most team leaders started looking at these previously unquestioned assumptions differently and learned to (1) ascribe more power and autonomy to team members to introduce their ideas, and (2) steer and structure the direction of the discussions, exerting less control over the outcomes. In this, leaders describe themselves as currently being more intentionally passive and patient, or more 'laid back' in their approach, and also as being more process-oriented and supportive, rather than defensive and guarded. Numerous statements confirm this transformation in the team managers, including: 'I ask more questions now, rather than sharing my own knowledge and ideas', 'I am less prominently present, sit in the back and let the team discuss', 'I was working too hard, now I try to relax more', 'I came with a solution too quickly, now I wait'.

Video-based aspects
The increased practical consciousness was transformative in the sense that it allowed the rest of the team to be more encouraged and acknowledged in their own roles and responsibilities. The idea of the new working method, although this was not made explicit before, is that the team works efficiently and responsibly without the dominant presence of a leader. In that sense, the team managers felt that video method in many ways contributed to making these values tangible, while also articulating and strengthening this Increased practical consciousness as reflected in leaders adapting their routines, including ways to deal with time, structure, and issues of autonomy in team meetings Comprehensive, rich and in-situ representation of practices and routines, including verbal as well as non-verbal actions and reactions Increased sense of unconscious motives, such as the need for self-validation and control The sequence of providing and discussing the feedback from peers, and discussing eye-openers together, invited participants to think about -and explore -the motives behind actions that are otherwise not discussed Increased discursive awareness reflected in a common language, including description of leadership roles The confrontation between video material and theoretical frames increased knowledgeability. Questions asked during the FGDs invited team leaders to negotiate the current and future practices with specific reference to their ideas -individual and collective -about the vision of the organization. This co-creative process of the method encouraged participants to discuss the theoretical framework underpinning various leadership styles and which approach might be preferred in the current context of youth protection Enhanced transformative capacity All elements combined led to a spiral of enhanced transformative capacity new leadership approach. The result described by team managers is that staff members now have a similar sort of mindset and feel more invited to engage. One team leader: Rather than talking about their practices (in which people discuss memories), video reflection confronted the participants with both non-verbal and verbal material, and moment-to-moment actions and reactions, which gave them deep insights into patterns of which they were otherwise unaware. Some of the participants' reflections about their revelations included how they positioned themselves in the physical space and what this meant to them as team leaders. Observations, for instance about whether they maintained eye contact with the team or focused on their laptop, about the status of the people that were seated next to them, etc. One participant said: 'I saw for instance that I can be very much engaged, and that I look at people quite intensely, of which I was not aware before. And that in itself I thought was quite fine. But I saw also, and my peers showed me, that me and the psychologist were sitting very closely to each other, and that it looked a bit as if we were forming a "block" [of expertise]. After that I decided to sit apart from [the psychologist]'. (TM-3) Most participants shared having experienced a sense of activation or elucidation of parts of their practical consciousness that was previously in the dark. It is important to understand how the multidimensional character of video contributed to this. An example was provided by one participant: 'Well, what I saw was a pitfall of mine, is that I was so much focused on the content, and although that is a power, I was too engaged in responding constantly. I thought, wow, if I would be my own manager, then I wouldn't like that. But only when I saw it back like this, I realized this.'  Most participants agreed that the video learning helped them become aware of specific intricate details of their practices. As such, the videos contributed to how they could address them, such as sitting differently or allowing others to monitor a laptop rather than doing it themselves.
More directly related to the structuring of the team meetings was 2484one peer's comment that the team leader took up too much time on exploring the problem, so there was not enough time to use this exploration for deciding on how to proceed. The team leader said that after this criticism, he experimented with some new routines in which there was more balance and even integration between the step of exploring the complexity of the problem and then using these insights for decision-making.

Discursive consciousness
Participants also reported that during the group meetings, they felt inspired to negotiate the various ways team leaders had previously managed their teams and have an in-depth discussion about the characteristics of leadership that should be prioritized in the organization's present and future context. This meant that the material that came up during the videos, as well as in the feedback, was confronted with the ideas people cherished about the collective vision of the organization ('every child safe'), as well as their ideas on what staff members most needed to function in accordance with the vision. For some participants, this was a crucial part of the learning process.
'Yes, for me it was important to understand from what implicit theoretical frame we manage a team. There are so many theoretical schools, but we are trying to learn how to manage a "learning organization". And about how to be a "learning manager", not a lot is discussed about that.' (TM-7) The learning process involved realizing that the vision regarding their leadership style is of crucial importance: '…, because it seems that people are not conscious of the fact that we do lead from a particular mental picture of what leading entails. And for a particular organization, we do need a distinct representation of a style. By discussing this together'. (TM-7) Developing a language and shared understanding together enhanced the team leaders' discursive consciousness, contributing to their learning and change process. During the reflection sessions, an important topic of discussion was the different roles a team manager could fulfil (or routinely fulfils) and how these could be symbolically represented. These roles can be seen as metaphors how leaders can play their role with a certain set of knowledge, skills and attitudes. Four main roles were defined, including: • The expert: The expert team manager is the one who gives all the answers and asks most of the questions, without leaving room for the team members to actively participate in team meetings. This results in a cycle in which the team members expect the team manager to give all the answers, with the result that they do not develop their own competencies. These team leaders received feedback during the peer review that this should not be their main role. Moreover, team leaders playing the expert role carried out the tasks of the family leaders, senior family manager or psychologist. The senior family manager, for instance, should be the one with the expertise in conversation techniques, yet the team manager carried out this task. • The minutes secretary: The minutes secretary is characterized by constantly using the computer and taking notes. These team leaders were taking notes of what was discussed and the decisions that were made during the basic team meeting. The feedback showed the surprised reactions of two of the team leaders, because they had no idea that their peers were interpreting their role in this way. • The facilitator: The facilitator is the team manager who actively stimulates the team members to participate during the basic team meeting. These team leaders are not only empowering but are also structuring the meetings by asking questions, and summarizing discussions. These team leaders were able to make use of the competencies and expertise of the team members at the meeting. • The decision-maker: While discussing a case, the decision-maker regularly decides what the next step should be, without asking the opinion of the senior supervisor, psychologist or the family manager. This may lead to team members feeling less responsible for their own decisions and choices regarding complex families and situations. If there was any uncertainty regarding a topic, the team manager made the final decision.
As stated earlier, the mission of the youth care protection agency is that every child is safe. From that perspective, one team manager argued that it is necessary to use all possible resources that help achieve the organization's mission and develop people. With that focus in mind, the facilitator role seems to be most appropriate, but each team manager has their own emphasis regarding the balance between the roles they play. In the team leaders' meetings, there is a lot of combined expertise and knowledge to tackle complex problems jointly. Stimulating participation, asking good questions and openness are key elements in generating the necessary information to address complex problems: 'What I try to do, is to ask better questions with the focus on the safety of the child. It is tough sometimes to keep doing that and sometimes people talk in circles with each other, but we have to steer the conversation always to the safety of the child and good questions help with that.' (TM-8)

Video aspects: increased knowledgeability through reflection and discussion
More discursive awareness of the different leadership roles made leaders feel better equipped to deal with diverse situations. The video-based learning method supported the learning process by encouraging participants to reflect on each other's leadership styles as reflected in the videos, and how these worked in in diverse leadership contexts. It helped them see how these different approaches were not entirely random, yet at the same time not well articulated. The video-based discussions stimulated the team to co-create knowledge together, generated by the gap between practice and understanding. The questions participants asked in these groups, such as 'why does this type of behaviour work or not in situation A, B and C', as well as 'what is it exactly that we are trying to achieve?', were instrumental in bringing more joint knowledgeability in the teams.
As part of this, what Giddens calls increased knowledgeability, there is a felt need to discuss with each other explicitly the most useful approaches to guiding their teams. Discussions with the team leaders in the reflection sessions met this need.

'One of the most important aspects of video-based reflection is the possibility to discuss the different videos with your peers.' (TM-10) 'You see a team leader being calm and steering the meeting in a direction and you wonder why is she doing that, what is the rationale and therefore it is important to talk with each other about it'. (TM-4)
We often assume that our behaviours present a certain norm, but the videos forced leaders to acknowledge the wide variety of possible choices in leading teams. Participants expressed their amazement regarding the differences observed; this is something they had never expected. Simultaneously, the differences, for example, between guiding the team meetings as a decision-maker, an expert, or facilitator, made it attractive to learn from each other and ask each other why they acted in certain ways. The videobased reflection method encouraged the team leaders to venture into a learning and a change process with their peers, for instance, about the different meanings of the concept of 'steering'.

'What kind of leadership style is necessary for a so-called learning organization? I don't think the concept steering is clear to everybody. If you mean by steering that not everybody goes on holiday at the same time, that is one way to look at it. But if you mean by steering to achieve that family managers improve their competencies, then that has a whole other meaning. Thus, we needed to develop a language between team leaders in order to understand each other.' (TM-7)
The comprehensiveness of the video-based learning method, including the sequence of providing and discussing feedback from others, and discussing eye-openers together, helped participants to embed their acquired knowledge in a shared culture of new norms and values. In that sense, the ALAR process, which arose as part of the video-learning process, made a significant contribution to increasing the team leaders' discursive awareness. Through discussing the practices of their colleagues and themselves, participants were encouraged to explain some of their previous routines (thus making their unconscious motives more tangible) and jointly negotiate changes.
As such, for most team leaders, what happened first is that they were able to explain what benefits were associated with retaining a sense of control and what made it challenging to let go of control. Some of the unconscious motives were closely linked to a sense of self-validation that comes from 'working hard'. Several participants felt that it was important to 'be relevant for the team', or 'to mean something'. Providing knowledgeable and experience-driven answers had been one of the ways to assure team leaders that they were valuable leaders because of their expertise. Two participants expressed these feelings: 'I guess I felt that it was important to save people, instead of allowing them to gain their own power. I used to think that "Well, I need to be of value for these people!".' (TM-10) 'It's also that you think, well I have also some expertise in the field and I happen to have a lot of answers myself to provide from that experience. It is weird to realize that it is not necessary.' (TM-1) The unconscious motive of 'being meaningful' led team leaders, as they described themselves, to be 'working too hard'. One participant said about this: 'Well, I can feel it in myself, that "Ooh I'm working too hard again and I'm trying to convince people". Then I know I'm a bit off the right path again. But this is a hardwired characteristic for me.' (TM-3) The participants felt that this 'overactive working attitude' could be the downside of a very effective and good quality they also did not want to lose. Thanks to the videos they realized that they do not need to lose this quality but that it could be complemented with other skills -that it is possible to find a balance, and that they have different ways to act, depending on the situation.

Enhanced transformative capacity
Finally, the results clearly reflect the interaction between what happens at the collective level (discursive consciousness) and the individual practice level (practical consciousness). The discourse evoked during the discussion groups helped bring to light various insights among the participants. They realized, for example, that by structuring the meetings differently they would have time to focus both on learning goals and also meet operational targets in time. At the individual level the team leaders experimented with these insights, gradually integrating newly acquired knowledge into their own leadership practices.
A few team leaders explained that they were asked about the way they were structuring the team discussions. They saw how other team leaders were having less problems with, for example, timekeeping. Although they were first hesitant or even apprehensive about following their peers' approaches, the group discussion increased the joint knowledgeability (e.g. the range of possibilities), which helped leaders navigate new styles of leadership. A team leader, for instance, explains how she restructured her team meeting by not starting with a question from the family manager but with a suggestion; something she picked up from her peers:

'Yes. And what makes a huge difference is that as the team leader, you do not always ask a question to the team, but share your intention. A question often leads to a counterquestion. If you put down an intention, it is more on point, you can steer the discussion more efficiently. One question is: would it be desirable to apply for an OTS [a court measure placing a child under supervision by youth protection] And one intention is: I apply for an OTS. And that is different approach, a family manager has a very clear position. With a question you have a thought, yes, but then my experience is that you sometimes get endless discussions. If you are a bit firmer and pointed about that and especially if it is based on the safety of the child (purpose of the organization), that is very effective'. (TM-4)
This integration of mainly skills and attitudes can be conceptualized as increased knowledgeability. Increased discursive awareness is part and evidence of this process of integration. The team leaders' interviews show many examples of how they kept on experimenting with smaller and more significant issues. Sometimes they discussed their challenges with the peer with whom they worked in the video reflection, but more often with their team members or others inside or outside the organization. All these learning activities led to increased knowledgeability, practical and discursive consciousness, and thus enhanced transformational capacity.

Discussion
In this study, we aimed to understand how a video-based reflection method could support an Action Learning and Action Research (ALAR) process to enhance transformative capacity among a group of leaders in a radically changing youth care organization.
As described in the introduction, video is a ubiquitous technological instrument used for reflection and learning and has been applied in various educational settings. The videobased reflection method used in this study builds on a strong tradition of interactive learning techniques, which have been developed since the information age, by adding aspects of reflective monitoring to the learning process. The current video-based reflection method combined aspects of representation with a co-creative process of deliberating new contextspecific constructs in order to stimulate learning among leaders.
The video-based reflection process included an ALAR-based methodology. Action learning generated outcomes in adapted leadership style, reflecting a greater ability to deal with tensions as well as to structure their team meetings. Action research helped to conceptualize these changes as increased transformational capacity through enhanced discursive awareness and insight in practical knowledge about routines. These study outcomes realized with the ALAR methodology are closely interwoven. In this discussion, we address the observed changes in leaders as a result of the video-based ALAR process, as well as some of the practical considerations involved with implementing an ALAR-based process in changing organizations.

Transformative capacity
Giddens' theory of transformative capacity was used to analyze the data and understand what leaders, about a year after the video-reflection intervention, brought up as tangible examples of change in their leadership practice. From this vantage point, the aspects of the method that were conducive to their increased transformative capacity were explored. From the in-depth interviews held with team leaders, it emerged strongly that the learning process had helped them expand their leadership capacities, particularly in letting go of control and empowering their team members. Empowerment is often associated with aspects of transformative leadership in organizations interested in engendering valuedriven practices in teams to respond more effectively to the complexities of contemporary work fields. Leaders reported that the video-based reflection method enabled them to see when they were likely to express dominance and control patterns, and learned to practice a different approach; one that involved more patience, appreciative inquiry and 'relaxed presence'.
In line with this, leaders learned to steer and structure the team meetings in a more context-sensitive manner. By watching the selected video footage together, and seeing how their peers' leadership styles diverged from their own, leaders were able to expand their range of strategies in response to diverse leadership situations. While initially most leaders would gravitate to one side of the control spectrum (either holding a tight grip or employing a laissez-faire approach), leaders learned to integrate aspects of both styles into their practice, and respond more flexibly and appropriately in their meetings. In other words, leaders became better-versed in using various techniques to bring empowerment, harmony and conflict-resolution into their teams. This is important because leaders often face situations that reflect a great amount of tension, particularly when the organization is in transition (Smith et al., 2017;Zheng et al., 2018). Various scholars have written about the so-called 'paradox theory' to provide solace to leaders who are commonly dealing with a 'messy, apparently unexplainable, and often seemingly irrational contemporary world' as well as the '….dynamic contradictions that come with every day decisions and activities in organizations and society' (K. Smith et al., 2017, p. 304). It is in such context that the leaders in this study were operating, and were appreciating the learning that was derived from the video-supported ALAR process. These outcomes were surprising as none of the leaders had received training in recognizing tensions prior to the video-based reflection process, nor was specific attention provided to elicit learning on this topic.

Higher-order learning through video-based reflection
A probable explanation for the increased transformative capacity could be found, first of all, in the dissonance created by video-based reflection, which pushes leaders to test and advance their own preconceptions (e.g. Reuker & Künzell, 2021;Rosaen et al., 2008). Without the video-based material, leaders rely solely on their memory to learn, whereas the videos enable them to view themselves and others, which helps them grow. In addition, our findings allude to the idea that the dissonance created, and the interaction encouraged in team discussions as part of the video-based reflection process, enhanced the discursive consciousness among leaders (Giddens, 1984). This type of learning is sometimes referred to also as 'higher-order learning', which is less often witnessed in traditional forms of video learning and instruction. The ALAR process, which with its cycles of action and reflection could be seen as a form of Reflexive Monitoring in Action (RMA), enabled leaders to imagine -and embody -a new leadership style. Combining video technology with a learning context that invites deep questioning and re-imagination of current practices has also been emphasized by other scholars (e.g. Koole, 2012). Open-ended questions and feedback from peers serve as a springboard for lively discussion and further reflection (Borup et al., 2012). Outcomes of this study also show the impact of representation (inviting viewers to witness their own practices as well as those presented by peers) and active inquiry to reflect critically on otherwise unquestioned and latent routines in participants. The rich visual material of people acting in-situ helped to show learners the reality of their own and others' practice, which then supported a process of reflection. These conducive aspects have been emphasized by other scholars, such Clarke et al. (2014) and Yung et al. (2007).
Even so, what is novel and important about this study is the specific context of organizational transformation in which video-based ALAR was applied. While some recent studies have begun to acknowledge the importance of inquiry and reflection to stimulate deep or 'higher-order learning' in students (learning to think, analyze and create), video-based reflection helped to encourage higher-order learning, in relation to developing leadership in line with an organizational vision. To our knowledge, this has not been done before.

Implementing a video-based ALAR process
As discussed above, the peer video-reflection approach is aimed at facilitating the team leaders' transformative learning. The concept of transformational learning is not focused so much on incremental or adaptive learning, but rather reflects learning that 'transforms problematic frames of reference -sets of fixed assumptions and expectations (habits of mind, meaning perspectives, mindsets) -to make them more inclusive, discriminating, open, reflective, and emotionally able to change' (Mezirow, 2003, p.58). Without the voluntary and, in many ways enthusiastic (see, participation of those involved, such dramatic shifts are unlikely to be realized. It requires the willingness of participants to expose themselves to being evaluated by others, which many people perceive as making themselves vulnerable. The fact that the participants agreed to it implies that they were ready to be evaluated by their peers, and that nobody dropped out, suggests that the process was encouraging and safe enough for them to continue (see also. Following Zuber-Skerritt (2002), we would suggest that an appropriate design and structure of a peer video-based reflection process leads to more successful uptake as it engenders a sense of comradeship in learning. The peer video-reflection methodology employed here has a very specific and structured way to inform discussions and create an atmosphere of openness towards feedback from peers. An edited video was produced containing all peers participating in the FGDs, 'eye openers', 'question marks,' and 'comments' structured along with core themes. This video presented at the start of the FGDs showed clips from individual leaders along with similar clips of their peers highlighting the specific and overall challenges. The findings show that the video-based method was sometimes seen as confrontational. Yet, by focusing firstly on positive aspects, which were then followed up with themes that were shared by all peers, reflecting on unifying challenges made them 'comrades in adversity', all seeking understanding, problem-solving and transformation of their leadership practices. Other scholars write about purposely choosing and presenting video clips that show examples of 'good practices' and essential 'lessons' (Yung et al., 2007), which then help participants to build knowledge together on 'what to do' (Gaudin & Chaliès, 2015). Showing leadership examples in various team meeting situations raises consciousness among team leaders, without demotivating them. This is an important finding because it suggests that the video-based reflection approach leads not only to new leadership practices but also to a new culture of higher-order learning. The interviews substantiate this suggestion because they include various examples of how the team leaders continued to experiment with new ideas. The team leaders regularly discussed these experiments with their peers, but more often with their team members and others they trusted inside and outside the organization. The resulting spiral of increased knowledgeability of the team leaders led to a continuous process of enhanced transformational power, which the team leaders used to adapt the way they led the case discussions in team meetings.
Finally, while speculating on what drives an appropriate atmosphere for video-based ALAR, the role of the researcher merits specific mention. Indeed, an important part of the ALAR approach is the research component, which distinguishes the process from strictly action-oriented learning. The primary researcher's role involves more than the traditional interview techniques and observations. In ALAR, the researcher mediates actively, and adopts a critical learning approach, asking questions which provoke reflection and discussion, and helps conceptualize what is being learned in the interviews and discussion groups. The researcher intervenes with their research, and the abstractions made during the ALAR process (e.g. through cutting and editing the video material for each group discussion). In this, the researcher is involved in more than just capturing the learning process and is actually part of it. According to Postholm (2013), this requires that the researcher have some expertise that makes them competent to lead the action research process, such as the ability to adapt to the developing learning process and act in relation to relevant learning points for the study participants. According to Postholm (2013, p.517), such researchers should accept that they do 'not have complete control over the process and what happens, but rather must accept surprises and listen with patience, and be open, creative and responsive'. According to Ghaye et al. (2008), it means that the researcher has the ability to appreciate what is important and see how the future can be developed from the present.
Their forward-looking attitude as well as ability to envisage possible solutions and their respective benefits, is what enables the ALAR process.
While a constraint on replicating and upscaling ALAR research could be time investment, technical expertise and research capacity, the methodology could also be seen as a cost-effective tool to foster learning in a changing organization. Whereas standardized workshops often lack context-sensitivity in training staff members, and thus might create more problems for organizational leaders in recognizing how to apply lessons learned in their dynamic, everyday work setting, the method we have presented is found to have sustained resonance among leaders. Lacerenza et al., 2017) explain that organizations invest a lot in leadership programs, but that their success is limited due to mismatch of needs and how content is provided. Lacerenza et al. (2017, p.1670, e.g.: 'A training program might emphasize transactional leadership style behaviors, which may not fit an organizational culture that values a transformational leadership style'). A video-based ALAR process, led by a component mediator, could thus be more useful and efficient as a method for organizations that are dealing with complexity and uncertainty of outcomes, thus avoiding extra 'remedial' costs later on.

Strengths and limitations
The study's limitations are that we have yet to assess the impact on the family managers' activities, on the practices of staff members and the actual impact of their work on child safety. This could be an important topic for future research. Secondly, an important source of data consisted of interviews with team managers, possibly resulting in self-reporting bias. An effort was made to validate the interviewee's responses by triangulating the data derived from interviews, the video footage, the recorded FGDs and the field notes.

Conclusion
The different phases in the video-based ALAR approach, the interviews seven months later, and the lens of structuration theory to analyze the data, helped to gain detailed insights into how team leaders perceived, interpreted and integrated new ideas about leadership into their own practice. We depicted the increased knowledgeability of the team leaders which resulted in a continuous process of enhanced transformational capacity. The methodology employed in this study might be particularly useful for organizations on the brink of, or going through, radical transformation in order to better deal with complexity.