Swedish teachers’ and school leaders’ rebellion groups on Facebook- collective formations and administrators as gatekeepers

ABSTRACT This article explores Swedish teachers’ and school leaders’ Facebook rebellion groups as a medium where professional needs and actions can be formulated. Data consist of interviews with administrators representing the rebellion groups. Based on a theoretical perspective of teacher agency we searched for experiences and visions related to contextual aspects expressed in the groups. The results indicate a common experience of lacking resources among the groups. Through a well-balanced agency, they search for support among politicians in their striving to improve working conditions. However, the groups might develop different conditions for participating due to the formation of group-specific conversational climate about what is desirable and possible to post and discuss. Discussions on work-related issues might certainly be affected and the rebellion groups might contribute to preserve the specificity of each level in the school system. A split in the teachers’ rebellion group indicates virtual relationships between professionals are easy to give up in favour of new formations, making the virtual groups vulnerable. The results raise the question about changing cultural aspects and what it means to be a professional teacher/school leader. Will teacher agency in the future be dependent on participation in such social media groups?


Introduction
This study explores a relatively new phenomenon in Swedish contemporary education context, namely teachers' and school leaders' rebellion (uppror, in Swedish) groups on Facebook. The phenomenon as such is far from unique and during the last years an increased amount of rebellion groups have appeared in social media (Khazraee & Novak, 2018). Some of these have focused on broad democracy issues such as women's rights or specific political issues, thereby reaching a large amount of people sharing discussions on topics that would be much sensitive in real life, for example to wear a veil or not. Teachers, school leaders and other welfare professions seem to follow this trend, perhaps as social media provides an easy and highly visible site to become active agents in discussions of their own professional concerns. However, this phenomenon raises a number of crucial questions where we not yet know the answers. Also, the communication involved in the social media sites involves a great deal of uncertainty, such as who can read and spread the posts or how a post can be both enlarged and reduced, as described by research in social and digital media (Boyd, 2010;Khazraee & Novak, 2018). Such issues are hardly reflected upon in Facebook groups of educational professionals. Rather it is what Liljekvist, Randahl, van Bommel, and Olin-Scheller (2020) express as a possibility to share experiences and to get quick advice.
In this article, we understand the phenomenon of rebellion groups in the broader context of how conditions for teachers and school leaders have changed over time, which has been thoroughly reported and discussed in educational research conducted during the last decades. There is no doubt there is a need to discuss work-related issues, such as recruitment of teachers, which tasks should be included in teachers work and other circumstances surrounding their every-day efforts. What is far from well-known, however, is what kind of work-related issues can be discussed through the currently upcoming Facebook rebellion groups and what conditions shape such groups. For example, in previous research it has been reported that moderating such groups often hold a tricky dual role, described by Hillman, Lundin, Rensfeldt, Lantz-Andersson, and Peterson (2021) as balancing between supporting participation and maintaining community standards. As these groups gather thousands of members reading, sharing and commenting several posts every day on work-related issues we might consider the Facebook rebellion groups to be in a position as powerful actors who give voice to educational professionals in new ways.
Our focus in this study is on the Facebook rebellion groups that gather teachers and school leaders. Empirically we have chosen rebellion groups representing three different levels of the Swedish educational system, namely preschool teachers, teachers and school leaders. Even though these different professions share several concerns, we claim missions and traditions differ which most probably will affect the formation of each rebellion group. In order to understand the Facebook rebellion groups per se and their possible contribution as sites for discussions of work-related issues, our data consist of information from the groups' Facebook pages and interviews with representatives of their administrators. Against this background, the aim of this article is to examine teachers' and school leaders' rebellion groups on Facebook as a medium for these professionals to discuss different issues and to potentially formulate needs and actions. More specifically, our interest is directed towards the conditions for participating, not what is actually posted and commented on in the Facebook rebellion groups. To fulfill this aim the study is supported by a theoretical understanding of teacher agency as an ecological approach, holding a temporal view on the past, envisioned future and possible actions that can be taken in the present as well as how the administrators relate to different contextual aspects. The following questions guide the analysis: What experiences and what visions and goals lies behind these rebellion groups establishment on Facebook, and what contextual aspects do the administrators refer to when talking about their groups?
What conditions for participation and discussions are formed in each rebellion group?

Research about professional conditions in educational settings
The current situations related to the Swedish school settings the rebellion groups represent, show some of the clues to what drives their establishment. In preschool, we have noticed an increased number of staff lacking formal teacher education while at the same time the preschools pedagogical mission has been enhanced, placing an increased pressure on the preschool teachers (Hildén, 2021). In schools, the shortage of teachers is noticeable as well as the increased percentage of unqualified teachers. The working conditions in schools are holding more and more administrative missions and the teachers' workload has increased due to demands of documentation, turning the teacher into a "paper pedagogue" (Samuelsson, Östlin Brismark, & Löfgren, 2018). When looking at the conditions expressed by school leaders in Sweden they struggle to improve school outcomes as well as equity in education and how to ensure the organisational and social work environment for teachers in preschool and school (Lundström, Holm, & Erixon Arreman, 2017;Norqvist & Poromaa Isling, 2020).
That professional conditions have been challenged in many ways can be explained in terms of a thorough reformation of the Swedish educational system (Hardy, Rönnerman, & Beach, 2019). Reforms about independent schools and the possibility for parents to choose schools for their children (Carlgren & Klette, 2008) have led teachers to increasingly market their work towards a more consumer oriented teaching or preschool practice, to attract potential customers. Other reforms have come to force in order strengthen control and accountability (Bergh, 2015), but also initiatives to facilitate teachers' work and to increase their status. Two example of the latter is that teacher assistants quite recently were introduced as a way to unburden the teachers' administrational work and the so called advanced teacher reform (Bergh, Löfdahl Hultman & Englund, 2019), which aimed to give teachers an opportunity to invest in their career and increase their salary. However the intended effects of introducing teacher assistants does not seem to have been achieved (Lindqvist et al., 2020) and reforms that focus the individual teacher have been found to decrease collegial cohesion among teachers who work at the same workplace (Murray, 2020). Such perceived issues and burdens are long since addressed in ongoing discussions between employers, employees and teacher unions, suggesting that the need of the rebellion groups to be redundant. Nevertheless, there is a large amount of research showing that many experience that teacher unions fail to support their members and how teachers stress other aspect of their work than the unions do (Bascia, 2008;Krantz & Fritzén, 2021;Lilja, 2013).
Facebook groups is a media in which professionals themselves can decide the topic for work-related discussions. Most teachers' Facebook groups have no rebellion intentions, but serve as support and to exchange teaching experiences (Bergviken Rensfeldt, Hillman, & Selwyn, 2018;Liljekvist et al., 2020). Beside these supporting groups, during the last decade teachers have founded several groups on social media also to protest or create resistance against different work-related aspects. Brickner (2016) studied two Twitter campaigns that were dissenting against a public narrative of the financial burden related to caring aspects in education. The twitter campaigns were successful as they pushed teachers later on to be more engaged in other issues of educational policy. A new genre of teachers' online activism and resistance is publically posted resignation letters (Hadley Dunn, 2018). Such letters contribute knowledge about bad working conditions even if posted by those who already have left their profession. Another example is the Norwegian preschool rebellion group, which in a few days in 2016 gathered more than 6,000 members in a successful protest against a parliamentary bill on language standards in the Norwegian preschool. The protest contributed to the Parliaments' withdrawal of the proposal (Fosse, Lange, Lossius, & Meaney, 2018). Among the successes one must also be aware of possible risks, as argued by Khazraee and Novak (2018), a general risk is that the goals of these groups often hold issues that no one will argue about, based on a desire to reach a large amount of members.
Our previous knowledge about conditions and demands on the educational professionals, only briefly referred to in this article, comes from a methodological and theoretical rich and broad field of international research over several decades. Results are often directed towards policy makers in an effort to contribute change. However, similar knowledge about professionals' engagement in work-related issues in social media is hitherto limited, and even more limited when it comes to research on groups of resistance and rebellion groups. The large amount of professionals engaged in workrelated issues in social media calls for a need to include research results from studies of educational professionals' experiences expressed in social media in the research field state-of-the-art. In this article, it is our intention to contribute to this discussion.

Empirical data
Data consist of descriptions from each rebellion group's Facebook pages and interviews with their administrators: The Preschool Rebellion, The Teacher Rebellion, The School Leader Rebellion and The School Rebellion. The groups differ in several aspects as outlined in Table 1. In February 2021, the School Rebellion, appeared on Facebook as a new group that within a month gathered more than 3,000 members simultaneously as the Teacher Rebellion lost around 500 members. To gather data, as a first step we read each group's Facebook page to get a hint of the groups' structure and goals. These pages hold short introductions about the group interests, what they strive to achieve, if they are private (one must apply for membership and only members can post and read post) or public (anyone can be a member and everyone on Facebook can read posts) and a set of rules from the administrators. The School Leader rebellion, however does not have any explicit rules on their Facebook page while the Teacher Rebellion explicitly state the administrators interpretive precedence and possibilities to, within the group of administrators, reject posts and exclude members. Second, building on the knowledge from the first step, we developed an interview guide addressing questions on how and why the groups were established, to get an in-depth focus on what they want to achieve and with what means they strive to reach their visions. In December 2019 we conducted qualitative semi-structured interviews with administrators from PR, TR and SLR to get a deeper understanding on both the single rebellion group, and on the establishment of teacher rebellion groups as a phenomenon per se. The four interviews lasted for about one hour each and were carried out in various ways; one physical meeting with an administrator of TR, one video talk with two administrators of PR, one telephone interview with one of the administrators of SLR. More than a year later, in February 2021, when a new teacher rebellion group, closely connected to TR had started, we conducted the fourth interview, via video link, with one of the administrators for SR. The focus in this interview was mostly on why a new teacher rebellion started and what differed in what they wanted to achieve. The first three interviews were transcribed verbatim and hold around 7,500-8,500 words each. The last interview was partly transcribed and included parts that gave a complementary picture of this new group. Quotations presented in the results are translations from Swedish to English in which we strive to be as authentic as possible.

Ethical considerations
All rebellion groups have a Facebook page in which anyone can find data concerning aims, rules for participating, the administrators' roles and, if needed, how to apply for a membership. All of us researchers became members in PR, TR and SLR and finally joined the SR. When entering the groups we declared our mission to be members only as researchers. As members we came to know which persons were administrators in each group and how to contact them via messenger or email. All administrators in each group were repeatedly informed of our research role in the group simultaneously as they were asked to participate or to select some of them to represent the group. All groups selected one or two of their administrators. All administrators participated in the interviews voluntarily. They were informed about their rights to withdraw as well as how personal data are stored according to GDPR. We informed about the possibilities to be recognised as the number of administrators to sample from are quite small. During analysis this made us consider if some strong statements from TR, e.g. on "democratic dictatorship" were suitable and ethically sound to present, but ended up it would not cause any harm for the participant, or the rebellion group, as it was uttered several times during the interview. Later on this concept was referred to in the interview with the administrator from SR, and we considered it to be a well-known attribute of TR. The study is part of a project that has been reviewed by the local ethical board at Karlstad University in Sweden.

Teacher agency as a theoretical perspective
We rely on a theoretical understanding of teacher agency (Biesta, Priestly, & Robinson, 2015;Priestley, Biesta, & Robinson, 2015a, 2015b stating that teachers' agency, their capability to act in the present and to express ideas and visions for the future are "invariably largely rooted in teachers" prior experiences' (Priestley et al., 2015a, p. 6). The notion of collective agency is discussed by Hökkä, Vähäsantanen, and Mahlakaarto (2017) as holding a close relationship with individual agency. They argue activities related to work practices start from individual initiatives and expand to become collective outcomes via shared endeavours. Hökka et al. state "collective agency can be manifested in the collective initiatives of a group of employees to develop new work practices, or their collective efforts to negotiate a new shared understanding of themselves as a professional group amid external challenges" (p. 38). We rely on their reasoning when regarding experiences to be individual as well as collective and the collective dimension of experiences, discussed and built up among educational professionals in preschool and school, to form the basis for what is controlled and staged by those who administer the rebellion groups. Agency may be about enriching the current school discourses, or as stated by Priestley, Edwards, Priestley, and Miller (2012), to challenge such discourses. Previous research on work-related issues as presented above, indicates a number of collective experiences -often negative ones -which contribute the need to challenge and set new goals for conditions in school and preschool. Research has stressed the awareness of the active and agentic role of teachers as change agents in professional development. Imants and Van der Wal (2020) argue '. . . agency is about individuals or collectives who are interacting with and within specific contexts' (p. 2. Italics in original). Being in a context where many ideas and actions take place, like the rebellion Facebook groups, creates a richer base of collective experience (Priestley et al., 2015a) contributing to the administrators' agency. Our basic assumption is that such experiences relate to their specific educational setting making the conditions in each Facebook group depending on the administrators' agency, which means their possibilities to act and form their visions in the surrounding context.
Relating to the ecological approach of the theory means we focus on cultural, structural and material aspects in the context the administrators refer to. These aspects are addressed as a part of an historical and recent development of the professional conditions (Priestley et al., 2015b). In our data, cultural aspects mean collective notions about what is considered to be reasonable and acceptable professional conditions for a teacher/school leader. Structural aspects deal with the social support educational professionals get from each other, their employers, teachers unions etc. Material aspects are financial resources, such as related to working time, teacher ratio, student equality or the physical school environment.
In this article, we focus on the administrators of the Facebook rebellion groups to act as agents, holding an overarching influence when establishing the Facebook rebellion groups and to share the collective experiences from the group they represent. Thereby we are interested in how the administrators relate traditions and visions to the ecological aspects in the education system when setting the current goals; when discussing what they want to achieve and with what means.

Analysis
Two related analytical grids have been used during analysis. The initial empirical interview data were read through several times searching for temporal expressions, such as the administrators' experiences, actions and visions with the establishment of the Facebook rebellion group. Further, these temporal dimensions were related to the material, cultural or structural aspects (Priestley et al., 2015b) when mapping grounds, ideas and visions in the groups. Results from the interviews are built up on the same structure showing experiences, actions or visions and how these temporal dimensions relate to the ecological approach in terms of cultural, social and material aspects. Data from the last interview follow the same analytical step and will be presented as a temporal description on what happened to one of the rebellion groups as the corona pandemic appeared in our society.

Results
The following results show how the rebellion groups started and how their administrators developed as different kind of gatekeepers. Further, how a common perceived lack of resources made the rebellion groups make use of different contextual aspects in the formation of their respective agenda.
All quotations are from the interviews and presented in italics.

Collective formations and administrators becoming gatekeepers
First, we present a brief background, based on the interviews with the administrators who have appointed themselves to administrate their specific rebellion group. They spend several hours each day outside working hours organising and controlling the memberships, desirable content and future direction of each group, actually making them work as gatekeepers. Being a gatekeeper means representing the rebellion group as a collective formation outwardly, however inwardly, it means to preserve each group's particularity and to visualise its specific concerns. SLR formed their rebellion to make school leaders and their specific concerns visible and to be part of a broader welfare, not only representing the schools: We are thinking of gathering the whole welfare sector under the same umbrella. Then we are 1.3 million voters. Then the politicians must listen.
That's how we think. Most rebellions are very positive, they will continue in their present form, but it will become a platform for collaboration (SLR). The same striving is found in TR, by moving ahead as a joint force looking for a progression: We have to make a progression; we cannot remain at the level of the Teacher's March. We must move ahead towards the School March and then towards the Welfare March (TR). Related to such joint activities PR express a weaker position, as they perceive preschool issues to become invisible vis-a-vis TR: We have stated from the start that we are purely a preschool rebellion. TR has asked us to join the Welfare March in the fall, we said "No thanks". Then we were asked if we don't want to support them. Yes, we want to support but we feel we want to highlight the preschool, or we will always be behind (PR). This illustrates a common interest in actions to influence politics, but emphasise their specific interests. It is obvious that the rebellion groups place themselves in a surrounding societal context. Structural aspects such as social support from other groups seems necessary when TR and SLR strive to make schools and teachers part of broader welfare issues. Contrary, PR stress cultural aspects to preserve a particular preschool culture vis-a-vis school cultures in the education system as a whole. The collective formations and competing on group-specific demands show a need to find out what experiences were brought in when the groups were formed.

Experiences of lacking resources
When establishing the groups, the goals to achieve were quite implicit, though similar experiences have triggered establishing the rebellion group. Cutbacks, increasing demands and lack of time altogether lead to imbalance in the professional conditions. Related to our theoretical understanding, such experiences deal with material aspects and overall lacking resources.
PR tells about a situation when one manager described a coming scenario and the feelings occurring although the staff had pointed out the problems many times: At a meeting we were told from our manager we won't be able to hire temp teachers anymore and we will have to think about this and this. Then, I was so angry. Because we had mentioned it so many times. We had written in quality accounts as well, that we could not fulfil our mission because we are getting more and more to do. The staff already started to get sick because of the job and the kids might suffer (PR).
TR tells about many small parts that eventually become overwhelming: It is, above all, the shortage, that is mainly between demands and resources and cutbacks and how it has started to affect the working environment. It is, for example, that you remove assisting staff and teachers are expected to do all the tasks themselves. Some have to teach more, and much more such peripheral tasks, like supervision during breaks. So that, the workload increases, more pupils in each class (TR).
When PR and TR tell about how cutbacks affect their everyday working situation, SLR refer to a more complex situation as a system failure at the political level, affecting the whole school system: The problem is the cutbacks in school that have been going on for a long time now. A little bit every year. Now, there are some managers now, they cannot even afford to have one teacher in each class without exceeding their budget. Then it's weird. Because everyone understands, you can't cut anymore. After all, the marketisation, and profits in the welfare system drains the municipal school of resources, which is extremely problematic. Not all independent schools are bad, and we shouldn't get rid of them just like that, but it is bad for the school system, for the Swedish school system in general and therefore we need to get rid of it (SLR).
The administrators share experiences of lacking resources, of cutbacks and a worry for the future if you do nothing. Professional experience here is used as a unifying force for a vision of a school with better material conditions. While PR and TR express personal professional worries related to material aspects, SLR agree but broaden the reasoning to include structural aspects when blaming the reform of independents schools to be responsible of the cutbacks. The different cultural and structural traditions and responsibilities in each rebellion develop from their perspective, what they and the professionals they represent experience in their everyday work will lead them to develop different agendas, as outlined below.

Making politicians aware
The frustration caused by lack of resources seems to rest partly on an idea about misdirected education policy, partly on locally financial decisions, thereby directing the rebellion to act towards politicians.
In their first action, PR administrators entered the scene during a municipal meeting and told about the shortcomings in their setting: We had no idea when we entered the scene, that politicians were present. We thought it would be our managers. Well, one politician tried to interrupt us. It was lucky for us, because then we got attention. Afterwards, many people came and said; now that you have done this, how are you going to proceed now? We hadn't thought it would continue at all, maybe just some meetings with politicians and managers. Then we just said, well we start a Facebook group (PR).
PR tells, through different actions, about to reach, talk to and make politicians reflect. They invited politicians to shadow the preschool staff during an ordinary day in preschool. Searching for social support in the societal context, politicians are very much a part of the structural aspects PR involve in their rebellion.
We notice a strong wish to find solutions based on consensus between rebellions and politicians. TRs' entrance to the arena was to start the rebellion group before the 2018 Swedish election in order to influence politicians: This rebellion was formed in 2018 just before the election, with the purpose to influence election issues. School related issues were highlighted, and many politicians said many good things. Then, after the election, it ended. That's how rebellions work. They die if nothing happens. I don't just want to sit and do nothing. So then we started the Teachers' March and that's how we got the rebellion active again.
The Teachers' March, in which many school leaders also participated, became the start for SLR, as a way to organise the school leaders. Besides debating the lack of resources in articles, politicians at both local and central level are subject to the SLR reminders: We try, after all, we have written some debate articles. We stress things locally. Like pepping each other and daring to raise the discussion in the various municipal committees, and managers meeting. We have met the Minister of Education, so we try to address directly to politicians, and we will meet with the parliament's education committee. We try to get direct contact with politicians, in parallel as we try to form opinion and carry on activism in our municipalities, as far as possible, on several levels simultaneously (SLR).
The groups share a faith in formal democracy as recurrently turning to politicians hoping the perceived problems can get attention. Even though not directly stated as a solution, we notice actions holding an implicit understanding of politicians' responsibility and power to change. Agency is expressed through involvement activating social resources and connections to make change happen.

Balancing their agency in the present
The rebellion groups are themselves part of the same structural aspects that supports other in the societal context and from which they need support to be able to make use of their agency. To do so they need to balance their agency in the present in order to avoid offending someone.
For example, SLR (as quoted earlier) are eager to tell about negative consequences derived from the reform of independent schools, but being careful that the colleagues who work there should not feel offended: Many of the principals, who are members of the rebellion group are working in independent schools themselves. So they get pretty upset about these ideas about equality. I do not mean that independent schools shouldn't remain, I don't think it's bad, but part of a system that is bad (SRL). Thereby SLR manage to balance between those who are directly involved and those affected by a certain policy, trying to make everyone satisfied.
PR let us know how they changed one of their goals on equality to fit irrespective of political rule: We want the preschool to form a safe ground for children. Previously we used to write about a preschool equal for all. Now we want our rebellion to work for achieving equal quality in preschool. It should not matter what political party, or parties, that govern in a municipality. What is written must be what is good for the preschool. One doesn't has to keep on changing; now we do this, and after next election we change to that. We want the preschool to be the same alwas, so you can rely on our values (PR). By stressing this structural aspect, PR avoid future battles with politicians who might come up with other ideas on equality, and can ensure its members will be pleased with a neutral formulation resting on neither left nor right wing policy.
TR make use of the same structural aspect when telling how to balance between showing financial injustices, yet not worrying other groups of people in the surrounding societal context. We want to influence politicians in general, because in this fights, it can be parents who are fighting against teachers, teachers who are fighting against leisure time centre, and teachers who are fighting against school management. There are so many different intrigues really, and we need to direct it towards those who decide on our budgets. The municipal politicians have the main responsibility that we try to influence (TR). The TR act of balancing show knowledge of how certain local policy needs to be reacted towards by the rebellion group to affect decisions meanwhile not offending, for example teachers or parents.
Altogether, we have shown how the rebellion groups struggle not to irritate any other group. Teachers and school leaders are civil servants not able in their profession to take a party political stance, but wrapping it in actions visible as "for the good sake". Thereby the rebellion groups, by explicitly avoid supporting some controversial policy, enacts as supporting other political ideas. We regard this use of structural aspects when shaping their social landscape also to show an awareness of a well-balanced agency, resting on experiences of disadvantages of making conflicts as well as visions of future good relation to other actors in the education system and to politicians. In a way the rebellion parts of the Facebook groups might be strengthening inwardly -however, outwardly it might look as if the rebellion is fading.

Talking, fighting and supporting
None of the groups get desired support from the teacher unions, thereby making their rebellion group to be understood as a 'supra union', taking care of the members needs and questions. A look at the preferable actions in each group we show actions preserving each group's particularity.
PRs actions relate to the tradition of working teams in preschool in which the teachers act supportive towards each other. As the most established group, PR sees no need to rush: We have heard democracy takes time, and we don't want any quick fix solutions (PR). When new members entering PR the norm of how to behave in the group is described as a layer of cotton wool as a supporting pad: It has been a fantastic support in the whole group, as we have worked it up. We get support when staying home having a fever or headache. It is like cotton wool around everyone. There is so much heart in the group (PR). When talking to TR, about what is allowed to call attention to in the group, the soft approach is not really what is preferred, rather more of activism like posts showing teachers who refuse to work overtime and who stop working for free. Maybe it is part of a (former) tradition with one single autonomous teacher in each classroom, making their own decisions. In TR, any member can post what they find suitable, but: . . . I usually describe this as a democratic dictatorship. We have a passionate administration who chooses what is reasonable. Such things are discussed in the admin-group, like small elements of democracy. There must be strong people who can decide, otherwise, it will be nothing. That is why we have a top management, but still it comes from below (TR). In SLR the specific intermediate position of being both employers and employees affects the school leaders to act in a more diplomatic way. They avoid identifiable criticism, instead staying on a macro level, making a situation recognisable and in the long run forming a more critical atmosphere: I think the managers are guarding their positions, especially if it is a small municipality. It can be very sensitive. That is why I thought if I start this, maybe in the end the managers dare to ask more critical questions (SLR).
Metaphorically, the PR-approach could be described in terms of the drip that erodes the stone, based in a deep faith for democratic processes. In contrast, the TR-approach is best described as more fiery and hot-tempered and therefore in need of a passionate leadership that can identify the orientation that keeps the level of engagement up among the growing number of members. The SLR-approach, finally, tries to negotiate in their intermediate domain and seeks to find a way to find impact without losing their confidence in the eye of the public. The actions taken relate to the cultural resources in our theoretical perspective and hold collective notions about what is considered to be reasonable and acceptable professional conditions for a professional preschool teacher, a teacher and a school leader. Acting in this direction might have possible consequences for the Facebook rebellion groups as narrowing both the professionals' agency and the possible topics for discussion.

One year later
One year after our interview TR split into two Facebook rebellion groups, TR and the new School rebellion (SR), focusing both on improving teachers' working conditions. Initially SR labelled themselves as Teacher rebellion 2.0, a clear signal to the TR on an updated approach. In our interview with one of the administrators of SR we were told TR came to focus too much on issues around the Corona pandemic, missing the real teacher-related issues. Even though TR pleaded for activism and used a democratic dictatorship, they were later on criticised for having too much bureaucracy and to having no activities. Contrary, SR actions were to publish a number of articles in a few months in daily press and on its own Facebook page. According to SR, TR came to be more like the teacher unions, with endless negotiations, like a ferry difficult to navigate, while the new SR regard themselves as: . . . a slightly fast motorboat that can turn quickly, i.e. handle questions quickly (SR). SR regard themselves as those who really give voice to the teachers, making us suggest the structural resources in terms of support from colleagues to be an important but vulnerable link when building the social landscape around the rebellion groups. The balancing of agency is no longer valued. Rather than striving for good relationships with politicians, the administrator from SR says: We want to protect our issues, whether they are uncomfortable or not. We want to help build a pride in the profession. We know the school better than politicians and unions (SR). It might seem a bit weird that SR, who plead for a unified corps of teachers being the ones who contributed the split by starting a new rebellion group. However, when a group of members won't get support for their desirable posts in the group, there is a risk that the rebellion group will split and new sub-groups appear, and that probably will damage rather than strengthen the common fight for improvements. SR pleads for fewer rules to follow and for less fixed content to be discussed. The administrator ends the interview stating: Those who do not want to follow us and participate can only choose another group, rather than tightening the conditions for participating (SR). This kind of loose relationship needs to be taken into account in discussions about the duration and importance of Facebook rebellion groups as actors on the education policy arena.

Conclusions
Summarising our results, we have found the rebellion groups all in general terms express a wish to achieve improvements in the Swedish education system and relate such improvements to their particular group of teachers, children/students or school leaders. The establishment of the School Rebellion indicate the groups are still in a phase of shaping their social landscape, establishing and recruiting members and defining their counterpart. This can be the tactical solution, referred to by Khazraee and Novak (2018), of reaching more members and being able to present a group of many members and thus a massive rebellion will be considered in future real life actions. However, virtual relations between professionals seem easy to give up in favour of new collective formations making the rebellions, in spite of thousands of members, vulnerable and in a short term weak. In the longer term, they may still be forums that contribute to increased mobilisation and strengthening of individuals daring to raise their voice in their own workplace.
Experiences and visions expressed by the group administrators differ and thereby the formation of conditions for participating in each rebellion group. Talking, fighting and supporting means the group of members within each rebellion group will meet a group-specific conversational climate and a norm about what is desirable and possible to post, certainly affecting discussions on work-related issues. The division between several Facebook groups will probably preserve the specificity of each level in the school system. On the one hand, this might make it harder for future policymakers to reach success in formulating comprehensive reforms, rather continuing to challenge professional conditions as shown in previous research (Hardy et al., 2019). On the other hand, the good side of it can be that specificities and different needs between levels better can come to the fore. There might also be difficulties to overcome in teacher unions striving towards a united trade union, contribute continuing failing in this respect (c.f. Bascia, 2008;Krantz & Fritzén, 2021;Lilja, 2013). In our introduction we considered the Facebook rebellion groups as possible powerful actors giving voice to educational professionals' in a new way. Based on our interviews with the administrators, we are no longer convinced. Vulnerability might harm the powerfulness while the preserving processes will provide professionals with the same old voice, even if the arena is new.
The rebellion groups take little and different account of democratic aspects when setting rules for the group. Acting as gatekeepers in social media actually means a few individuals have the power to decide about membership and affect the content of the rebellion groups (Boyd, 2010). Such in real life (IRL) processes of appointing leaders in a democratic way seem to be lacking in Facebook groups where the gatekeepers more or less are the founders of the group that remain as administrators and leaders for tens of thousands of members. Thereby the connections between Facebook rebellion groups holding aspects of a self-initiated and perhaps pleasant forms of power, versus the IRL staff meetings in which the immediate issues have to be democratically negotiated and dealt with is visualised. Previous research has shown that gatekeepers use Facebook groups to highlight their own interests and neglect others (Boyd, 2010) and how moderating such groups involve a balance between monitoring, facilitation, and expert membership (Hillman et al., 2021). Part of such aspects contributed the split of teachers' rebellion group, indicating some of the negative effects of self-initiated gatekeepers. We have also seen that the rebellion groups, while being displeased with TUs engagement, start to act like "supra unions" themselves, trying to support and solve practical problems among their members which will be even more problematic when self-initiated gatekeepers represent these kind of "supra unions".
Our use of teacher agency as a theoretical perspective has been useful so far to get sight of experiences used in the formation processes and how contextual aspects are stressed when goals and means are visualised. At first sight, virtual groups might be seen as distanced from real life. Contrary, our results show they are to a large extent part of the surrounding society, for example through striving for broader welfare alliances. The rebellion groups have, in different ways, approached politicians, which can be interpreted as an overall faith in political solutions to bring change. However, within the SR, such faith is replaced with a statement that teachers know the school better than politicians. If the SR agency in this respect will have success or if the other rebellions balancing their agency will reach success in the long term is an issue for further research. However, it raises a question as to whether these Facebook rebellion groups are changing the cultural aspects of what it means to be an educational professional. Will teacher agency in the future be dependent on participation or not in such groups?