Changing the gender balance in preschools: an analysis of active work carried out by seven Swedish municipalities

ABSTRACT The focus of this study is to examine seven municipalities that, in different ways, are working to recruit more men to work in preschools. The aim of the study is to identify different ways of working to recruit more men to work in preschools and to effect a change in the gender imbalance among the working staff in preschools. Relational agency is used as a conceptual tool in the analysis together with theories of organisational learning and change. The result is that municipalities use initiatives, one-dimensional processes or multi-dimensional processes when organising the change. These three ways of working characterise the work carried out. The work performed can in some cases be a mixture of all three, or at least comprise the first two—initiatives and one-dimensional processes. This shows how municipalities address the issue differently and the fact that the type of agency they take and are prepared to take—or the conditions needed to succeed—all differ. There is a great deal of variation in the work performed across the municipalities. One can discuss whether the municipalities more committed to change are also those more willing to have multi-dimensional processes.


Introduction
There is a great deal of international research concerning men in preschools (see Åberg & Hedlin, 2015;Barnard et al., 2000;Brody, 2015;Brownhill, Warin, & Wernersson, 2016;Cooney & Bittner, 2001;Cushman, 2005;Havung, 2000;Heikkilä, 2016;King, 1998;Mistry & Sood, 2013;Peeters, 2007;Stroud, Smith, Ealy, & Hurst, 2000;Sumsion, 2000). Several studies (e.g. Martino, 2008;Sargent, 2005) highlight how men who are either beginning work or who already work in preschools wish to work under the same conditions as their female colleagues. They want to contribute to the quality of preschools as professional employees rather than simply be present in preschools because of their gender, even though academic discussion around "educare" show how complex the arena of ECE is for men (Warin, 2014). Many municipalities also want more men to work in preschools. The lack of male teachers in preschools is seen as problematic in different ways by different actors, whereas having a group in a clear minority position can generally be said to present a risk of excluding that group from actively contributing to the work performed. Even though the number of men working in preschools is very low, and despite the fact that some initiatives taken to change that number have been taken (Johannesen, 2010), the issue of men working in preschools is wider than increasing the numbers of men working there. This is taken up by Peeters, Rohrmann, and Emilsen (2015). This article focuses on the strategies used by a number of municipalities to achieve the goal of recruiting more men into preschools.
There is a view that men should be male role models in preschools, and it is not always realised that this entails the need to find a certain type of man with a certain set of skills. (cf. Brownhill, 2015;Pirard, Schoenmaeckers, & Camus, 2015). In a study by Cushman (2008), 250 head teachers in New Zealand were interviewed. They were asked about what they mean when they use the term "male role model", with Cushman wondering what head teachers want when they say this. Her results showed that head teachers had positive associations with men who displayed "recognisable" masculinityheterosexual, rugby-playing men with the right attributes. They were strong, trustworthy, humorous and did not make the head teachers feel insecure (see also Francis & Skelton, 2001;Skelton, 2009).
In their study, McGrath and Sinclair (2013) highlight how male teachers can be role models for both boys and girls and how this provides social advantages to both groups. They also emphasise parental viewpoints and show how positive parents are in believing that men should also be role models for girls.
Based on his study, Haase (2008) believes that men who already work in preschools need to be supported in order to understand gender, power and structures so as to visualise their own role in working with children in preschools (Martino, Lingard, & Mills, 2004;Roulston & Mills, 2000;Smith, 2007). He believes that many men who work in preschools have a great knowledge gap about the context to which they belong. This becomes apparent in practice, as many distance themselves from what the women do and what might be associated with femininity in preschools. A number of men have no problems with this; however; these men can also reflectively discuss their role in preschool (Haase, 2008).
The former Swedish government highlighted the issue of gender imbalance among personnel in preschools (The Government/the Ministry of Education and Research, U2011/7067/S, U2011/263/S). The three political arguments that were used to start a number of initiatives so as to change this imbalance were, firstly, that children should see that both women and men can work in preschools and in this way the preschool as a societal institution would be a "role model" in itself for a more gender-equal society. A second argument used was that there is indirect gender discrimination, since men cannot freely choose to work in preschools because of the stereotypical assumptions concerning what women and men should choose as work (at the same time, there was a survey presented showing that around 37% of 16-28-year-old men would choose to work in the welfare sector, which, for instance, could mean working in preschools). The third argument was that there is a great lack of educated preschool teachers, and when doing long-term work to change attitudes towards working with small children, more men could be attracted back to preschool if they were already qualified and had chosen to work somewhere else, or to qualify to work in preschool.
The aim of this study is to identify the different ways of working in seven Swedish municipalities, when the municipalities actively decide to try to recruit more men to work in preschools and thereby try to change a gender imbalance among the working personnel in preschools. The aim is to elaborate on municipal agency as a specific process within this work.
The research questions that are in focus for this interactive study include: • How do municipalities work in order to recruit more men to work in preschools?
• How can this work be described?
• How can municipal agency be understood in this work?

Theoretical framework
As Granberg and Ohlsson (2016) indicate, there is an array of methods to describe, analyse and understand organisations and change in organisations. The field of organisational, education research is characterised by an interdisciplinary approach (Granberg & Ohlsson, 2016) and several knowledge areas are required to understand the subject studied hererecruiting more men in preschools. In this interactive research study, municipal agency is elaborated on as a key concept in the research questions, following Edwards (2005Edwards ( , 2011 relational agency. Agency involves several aspects of social practice and here municipalities are viewed as one organisation that can react to, manage and strategically work on complex social issues and dilemmas in different ways. To what extent this is done depends on how a municipality manages its ability and power to act: for example, if a municipality chooses to react to a societal issue or follow more of a monitoring strategy. In relation to the subject of men in preschools, municipalities create municipal agency by making certain choices or by not reacting. Both ways involve a type of agency-to act or not to act. Understanding the matter, here men in preschools, and the organisational structures are significant to a municipal agency, which is something that will be highlighted later in the analysis. Here, Ohlsson's (2016) discussion on team learning is useful in relation to relational agency. Styhre (2014) describes organisational change processes as being internal ones, dependent on actions, agency and preparedness within the organisation. A smaller internal theme alone cannot take care of the entire change-municipal agency is needed in addition. Styhre (2014) believes that such institutional work with its origins in "agency" has to be seen in relation to the actual actions performed by individuals in institutional practice. For development or change to occur in such a practice, for example in an entire municipality or municipal preschool, the institutional practice needs to view the actions performed as a constant, ongoing process without a specific end-point: "to underline actions as the distributed and collective capacity available to question established truths and norms" (Styhre, 2014, p. 105-own translation).
Organisations can have a different "reception capacity" or "susceptibility" (Callerstig & Lindholm, 2011;Heikkilä & Hellman, 2016) to change. This can be compared with what Styhre (2014) describes as something that institutional work needs to take into consideration. Institutional contexts provide social actions that are not equally visible; however, institutional work is regarded as different preparedness that can balance out different needs, interests and institutional logic. Styhre furthermore emphasises the importance of not viewing institutions as exclusive agencies that "are in place and structure social actions"; instead, institutional work is "the total number of activities constantly needed to create, maintain or disrupt an institution" (p. 108). In relation to a change process, it can be viewed as work to change how municipal agency is performed and prepared on site and how the social actions that arise are made visible in institutional practice.
The concept of "relational agency" (Edwards, 2005(Edwards, , 2011 serves as a way to analytically zoom in on how the local work in the seven municipalities work can be understood as municipal agency. "Relational agency" can be understood as a concept that focuses on how individuals' expertise, experience and knowledge stand in relation to those of others. Expertise, experience and knowledge is revealed and shown in relation to the institution where it is performed (Styhre, 2014). Edwards (2005) argues that relational agency means "a capacity to align one's thought and actions with those of others in order to interpret problems of practice and to respond to those interpretations" (pp. 169-170), which means that it is about having the capacity to put together thoughts and actions with others who are also involved in resolving a dilemma or a task. This should be done, according to Edwards, in relation to the interpretation of the dilemma or the task at hand, in order, to the greatest extent possible, to reach a good solution.
Here, Edward's (2005Edward's ( , 2011 concepts and Styhre's (2014) ideas on agency form the idea of municipal agency related to preschools. Municipal agency is here operationalised through an analysis of how the seven municipalities that were studied act, react and perform change in order to recruit more men to preschools, which they have made a commitment to do. Through analysis of the data collected, three aspects of municipal agency are found. These three aspects are presented in the Results section.

One network of municipalities
The study includes seven municipalities (Swedish kommun), all included in one network specifically aiming at raising the number of men working in preschool. These municipalities have different reasons for participating in the network, and that has also been the starting point for the network-different municipalities will participate and enrich each other and, regardless of their origins, obtain new knowledge that can support the work that is ongoing or planned. Between 2014 and 2016, the network met five times. The content of the meetings varied and was entirely based on the needs of the network participants. The participants communicated their needs to the project management at SALAR (Swedish Association of Local Authorities and Regions), and speakers were enlisted or different workshops planned.
The aim of the communication network is threefold. One initial purpose of bringing the participants together from the municipalities has been to enable an exchange of experiences and ideas on development to promote the local work of recruiting more men into preschools, for which different operators participate and support the municipalities. Another aim has been to ensure, maintain and develop high-quality preschool organisations. A third aim has been to disseminate positive experiences and knowledge to other municipalities in Sweden.
Seven municipalities participated in the network. They were selected for inclusion in the network through SALAR and their experiences as an organisation regarding the subject of men in preschools. The municipalities are located all over Sweden, are of various sizes, have more or less worked with gender equality and have shown an interest in the subject and want to participate in the work of recruiting more men to work in preschools in their local context.
The municipalities have been made anonymous in this study and the participants who provided the quotations are assigned aliases. This is for ethical reasons, since these municipalities can otherwise be recognised relatively easily.

Interactive research as method
The study builds on a method known as interactive research (see Callerstig & Lindholm, 2011;Svensson et al., 2013). Interactive research can generally be said to actively involve participants and researchers in creating the analysis and the direction of the study together. Svensson and Brulin (2013, p. 23) believe that "sustainable development work that will produce long-term effects builds upon learning and reflection of previous experiences that provide an understanding of the wider context in which a project is included" (own translation). Interactive research can be a tool to achieve such an understanding. This means that the researchers follow the participants in their work and attempt to analytically describe and systematise what appears to take place, with the aim being that the researchers see their activities and work with fresh eyes and obtain theoretical knowledge from it. The analysis and reflections from the researcher are "tested" by the participants and by taking their reflections seriously.
The interactive research conducted in this study was performed at specific times, at network meetings and when paying visits to municipalities and through holding interviews with local participants in the network. Discussions from the network meetings have been documented and form background material for interactive research. The participating municipalities were visited on a number of occasions, and interviews were conducted during these visits. Interviews with municipal representatives took place on three occasions: autumn 2014, spring 2015 and early spring 2016.

Interviews and visits
Interviews were carried out with key persons who were involved in the local work and involved in recruiting more men to preschools. The municipalities identified the key persons themselves. A "key person" is someone who participates in the ongoing work, or who plays an important role locally in recruiting more men to preschools in the municipality (Bryman, 2011). The positions these key persons held were, for example, preschool teacher, preschool manager, local school manager, preschool administrative manager, gender equality officer or school politicians, which means a variety of positions on different level in the preschool organisation.
Visits to the municipalities focused on how the municipalities work in practice and what they do, or what they intend to do, to recruit more men into preschools. For example, this may be to change or use policy documents differently, use recruitment routines or identify and establish what the processes that affect the number of men in preschools look like in the different municipalities.
The interviews were semi-structured and qualitative. They aimed to capture the participants' understanding and reflections of the phenomenon and the local process (Kvale & Brinkmann, 2014). There were relatively few interview questions, thus allowing the opportunity to ask follow-up questions. This was a conscious decision, as the local work of recruiting more men into preschools differs and, therefore, required different approaches in the interviews. All interviews were recorded by a mobile phone, and downloaded to a computer immediately after the interviews were finished.
Before the interviews, the participants were informed about the Swedish Research Council research ethics codes (Vetenskapsrådet, 2011). This means that participants are guaranteed anonymity in the analysis and its presentation; they have the right to withdraw from the study without notice and explanation, and they also have the right to examine and change the transcript of their interview to better reflect their understanding of the current situation. All interview transcripts were sent to the respective participants for approval.
In 2016, 24 people participated in the interviews: 14 on-site interviews were conducted, 3 individual and 11 group interviews, with between 2 and 5 participants in the group. The individual interviews varied in length from between 25 and 60 min. The focus group interviews lasted from 45 min to 2 h.

Analysis method
The material was analysed in several stages. To begin with, all material was carefully read through several times with a focus on the research questions. Notes were made throughout the reading when the material reflected the study's central concepts. Information that deviated considerably from the ordinary was noted: anything that may have "stood out" in the material, or described a process or measure, opinion or perspective. At a later stage in the analysis, the researcher returned to these notes that dealt with anything that, in different ways, varied, either coherently or in isolation.
Certain parts of the material have been further reviewed in order to find patterns between the highlighted sections. This work led to the emergence of disparate and clear patterns, with some aspects being more obvious than others. These tentative patterns were then re-checked against the material. Subsequently, emerging patterns were categorised and described in the text as part of the municipalities' work and their approach.
The analysis work was explorative and focused on the materials, alongside the analytic concepts and research questions. It was impossible to determine the results or direction of the analysis after the various materials had been gathered (cf. Callerstig, 2014;Callerstig & Lindholm, 2011).
Part of the analysis process for interactive research is to let the participants in the study gain access to the tentative initial results and analyses. For this study, this was carried out in April 2016. Here, the analyses that had been completed up to this point were presented to the participants at a network meeting. Different issues with the material led to group discussions and questioning, as well as certain queries.

Results
The results presentation focuses on the aim of the study, which is to identify different ways of working in seven Swedish municipalities, when they actively decided to try to recruit more men to work in preschools and thereby tried to change a gender imbalance among the working personnel in preschools.
In the practical work that the municipalities are doing, different concrete initiatives have been taken. These initiatives were different meetings, school information on how to choose educational paths in order to produce the most appropriate education in preschool, or lectures on different aspects of gender equality for current staff and for preschool leaders. It was also about creating gender neutral brochures or web-based information on what it meant to work in preschools, and for some municipalities it was also about critically examining the recruitment process in order to find gender biased aspects of that process. Also included in these initiatives was how to retain men who already worked in preschools and to identify them as a minority group that perhaps needed attention.
It can be concluded concerning the different initiatives taken that, when a municipality worked more with this issue, it became clear that, even if the first idea was to work with recruitment processes, in order to be able to recruit more men, initiatives of another character were needed-such as attitude changing initiatives, and more long-term initiatives for change, such as school information before making a vocational choice.
The following section presents an analysis of the work aiming at elaborating and answering the first and second research questions: How do municipalities work in order to recruit more men to work in preschools? How can this work be described? The third research question (How can relational agency be understood in this work?) is also included in this section and explored further.
The analysis is divided into three main themes-initiatives, one-dimensional processes and multi-dimensional processes, as shown in Figure 1. The initiatives were described briefly above. It should be noted however, that none of the municipalities worked only by generating new initiatives. Rather, all municipalities included in the research tried to establish processes that, in varying degrees, would be embedded in the work of the municipality. The results show that the way of organising the work into processes is central to how the work is conducted and in what way the work is expressed by the participants as being successful. The analysis presents and discusses all of these themes: initiatives, one-dimensional processes and multi-dimensional processes. The presentation of the results starts with a section discussing how the work was organised in the municipalities, followed by a section presenting the themes mentioned. The results are summed up in the closing section of the results section.

Organisational conditions for the work
The organisation of recruiting more men to preschools is important in terms of the development of the work in the municipality (Heikkilä, 2016). This applies in the same way as for other municipal change processes; however, it also demands that the subject of more men in preschools is viewed as a change process, which is not a given. Some of the participating municipalities take the view that "it's just about recruiting".
Several of the participating municipalities have created internal organisations that are working on the issue. These are often organised into groups, or teams, where participants with different skills meet, which can, in Edwards's (2011) terms, be called how a boundary space is created, or how they appear. In the municipalities where this way of organising appears to work well there are representatives from both preschools (ideally, managers with a comprehensive mandate), HR and gender equality services in the group. It can also be said, using the analytical concepts relational agency and boundary space (Edwards, 2005(Edwards, , 2011, that these boundary spaces, the groups, are arenas for municipal agency to happen in practice.
Based on an analysis of how the participating municipalities have developed their internal organisations, it becomes clear that the following are important aspects for creating relational agency and team learning, and thereby create a basis for collective learning and possible change: clarity about who has the mandate and how far this reaches; having a budget; and having a consistent approach, with the same people working with the issue.

Creating processes for change for a gender balanced workforce
The following section provides examples of how the characteristics of the work to recruit more men to preschools have changed during the two years of the network, and it also describes how this has taken place. One of the tentative reasons for this is that several, though not all, of the municipalities have within their work and the way they organise their work, developed relational agency, and within that, team learning. This is explored here by some examples of how this is also expressed by the participants who were interviewed. It also addresses the third research question of this paper.
The results presented in this section show how the municipalities have gone from arranging and discussing an array of individual initiatives to progressing and developing different types of processes. An analysis of one interview suggested that the agency of the municipalities has become more complex, and the preparedness regarding the subject of more men in preschools has become more professional. The analysis of the characteristics of the changes in working can be illustrated in the following way.
The figure shows how, at the beginning of the work on the theme of bringing more men into preschools, the municipalities use individual activities or initiatives-often of an informative nature-to try to increase the number of men. It does not take long before this is deemed insufficient and the group or other key actors identify different kinds of processes that need to be initiated. As a starting point, these can be onedimensional, meaning that they are characterised as being fairly time-limited and preschool internal. Alternatively, the targets of the processes are predictable.
The analysis also shows that a process that is initiated can be multi-dimensional. A multi-dimensional process most often follows a one-dimensional one when the former has proven too limited for the ambitions of the municipality. In a multi-dimensional process, the final goal is seldom known in advance, even if planning for a process includes formulating a certain target and establishes suggestions for its attainment. Several unknown internal and external interested parties can be part of such a multidimensional process. The analysis also indicates that two different processes can be found in parallel in a municipality. However, the likelihood is that the number of processes decreases if a multi-dimensional process is initiated, as both individual activities and one-dimensional processes have been incorporated as a part of the multi-dimensional process. Within this, the analysis suggests that creating awareness of relational agency and also boundary spaces is part of the process.
In the following section, these three themes are studied in more detail and exemplified.

Initiatives
As already stated, initiatives are often informative. Timewise, they are relatively limited and there is seldom any discussion of continuation. Nor is there any exchange on how to understand the topic of men in preschool as anything other than at a very superficial level. One example is school information for grade 8, or a guest lecturer being invited to come and discuss a topic such as gender in preschool (it is important to remember that this analysis does not mean that school information or speakers are something negative or counter-productive; the aim is to highlight how municipal efforts can have the greatest possible impact). These issues are seldom followed up. One variant of a selective measure is when an individual employee does not share any knowledge they may have, thus having no impact on the organisation. This is one way to describe a selective measure: We have worked-tried to work-a bit more with the gender perspective and a little more with a norm-critical perspective, and I think there is a shortage of this gender perspective, even at university level education. I have been and trained all the preschool teachers once at least.-Jonna, Gender Equality Representative This quotation shows that the person taught on university courses at the local university, but is uncertain of the results or the conditions available. Perhaps a number of selective measures become symbolic actions, as was the case here, as the university said, "at least we tried".

One-dimensional processes
The analysis shows that one-dimensional processes often focus on preschool aspects and what the preschool itself can do in order to recruit more male teachers and personnel. These processes are often time-limited and likely to have a restricted budget. A one-dimensional process can be about establishing new routines for obtaining information from the criminal records database, initiating training measures for norm-criticism in the preschool, or finding new types of collaboration with universities regarding teaching practice placements (VFU) for students. One preschool describes its collaboration with a university as follows: It's in the works. We've decided that all the guys studying to become preschool teachers will come and visit me at some point or another guy who is also a mentor. They'll spend some of their VFU with us; they're going to make sure of it, that they'll be with us.-Victor, Preschool Teacher Together, the municipalities and universities have found a way for VFU work to promote men in preschools by allocating male mentors to male students during their VFU. Using the analytical concept relational agency can be a way of regarding this cooperation with the university as a way of trying to develop the work done to recruit more men. It can be analysed as a starting point for creating a multi-dimensional process including several actors to resolve a societal dilemma such as the gender imbalance in preschools.
The following quotation provides another example of when several processes take place simultaneously, but where Birgitta only talks about it as one. This is interpreted as an example of a situation where different areas of expertise are included in order to handle the issue of gender imbalance: But I feel that this thing with school holiday work placements is good; we're really happy about this. And it was quite simple. Since then, we've planned . . . on 5 March 1 we're doing with . . . is . . . what must not happen happens, about abuse and children in preschool. And . . . since the target group is preschool managers and municipal representatives . . . we'll bring in others too: Oskar Larsson and Lars Petterson 2 who work with crisis management in Fisken Municipality, so that´s the current situation.-Birgitta, Organisation Manager Here, the person interviewed describes two different processes, both of which are understood as a one-dimensional process, later linking up with the multi-dimensional. The municipality has set a date for an internal conference on safety and abuse in preschools and has invited and enlisted a wide range of interested parties in addition to preschool personnel. The need for the conference arose in the work with the theme of men in preschools and was based on how the municipality's internal group increased its repertoire for the theme. The preschool is still the focus. Even if it is not possible to completely predict the effects of the conference, one of the desired outcomes is an increase in knowledge and awareness.
Here, a preschool manager's quotation provides another example of how education in gender and norm-critique was arranged to increase awareness of gender issues among the staff, which is also a way of creating relational agency: Yeah, I mean, I think it's become clearer, this discussion in the workgroup, the approach. Like, you're more aware of the differences in approaches to girls and boys, I think. And at the same time, I know that this gender perspective is really important; we need to bring this into the organisation. It mustn't just be something specific, it has to be integrated. And I think that it's really, really different in different units.

-Anton, Preschool Manager
The staff member describes how there has been an interactive element in the education, and he sees this as a process for the preschool to view the activities and how the gender perspective can also increase quality at preschool.

Multi-dimensional processes
Multi-dimensional processes often involve several parts of the municipality and raise preschool matters to a general municipal level. This is unusual. These processes may be both internal, concerning only the municipality itself, and they can also be external, to include other participants, other organisations or authorities. Even though the preschool issue is central, different organisations in the municipality are seen as being involved in the process of bringing more men into preschools, and supporting and providing a mandate for the desired local effects.
A multi-dimensional process requires a high level of interaction and communication, and in the multi-dimensional process it is possible to discuss the organisational issues as relational agency and team learning within different boundary spaces. Such processes require areas of expertise to meet, to interact and to try to find new solutions to common dilemmas. The analysis also shows that it is good to have discussed the basis for the local work of recruiting more men into preschools in depth, as it clarifies the desired local effects. As several parties are involved in a multi-dimensional process, another important part of the process is to enlist support for its different aspects. The multi-dimensional process combines the findings and experiences from one or more one-dimensional processes, without really knowing where it will lead, because of this "fumbling around in the dark". Some examples of what can be viewed as multidimensional processes will now be presented.
One example of a multi-dimensional process is when one of the municipalities discovered that a large, male-dominated company was about to make several redundancies. The municipality was able to attend one of the major employees' meetings and provide information about the preschool as a workplace and what it is like to work there. Two male employees from the municipality's preschools provided information and concluded by giving details of a network of supply teachers and what it involves. One of the managers describes the result: It's about a man in his 50s who had never stepped into a preschool but was recruited and is now stuck: well, not stuck, but has a fixed-term contract as a substitute teacher in a preschool. He's good and suitable and, in addition, he's interested in training.-Charlotta,

HR Consultant
The man now works in a municipal preschool as a permanent substitute teacher and has undergone introductory training. The actions of this municipality are an example of a process that starts with nobody involved knowing the outcome and where there are internal and external parties who do not know each other. And having this experience means that they are continuing to think in this way, trying unconventional as well as more traditional ways to try to find more employees for their preschools.
Another example is of a municipality that has created a mentorship scheme between a higher education institution and the municipality in order to reduce the drop-out rate from the preschool teacher programme: And now we have this mentorship thing going on, great. We're all training to become mentors for new students, as of August 2016. The guys that have started university and are studying to become preschool teachers will get a mentor-one of us. The guys that work for the municipality are contributing. They'll be given a mentor straight away and we'll go there at the start of the course and introduce ourselves as personal mentors.-Victor,

Preschool Teacher
Here, there is an idea that men are role models for other men, compared with a previous idea that men in preschools provide male role models for the children, which was more related to what "a real man" is and does. The municipality has worked with a general contextual idea and created a multi-dimensional process that will continue for a long time, where the result of the process is unknown. There is an idea of inclusion, where men at municipal preschools will help to include men studying to become preschool teachers in a context where the status of men as a minority in preschools will not frighten them away.
Collaboration with the Swedish Public Employment Agency (Arbetsförmedlingen) is also an example of a multi-dimensional process. One municipal employee explains: So, I'm on my way to Arbetsförmedlingen: I'm going there at the end of March. I and another guy are going to meet jobseekers. They have, we sort of heard. . . quite a bad reputation that came from somewhere, that employment officers have told jobseekers that it's a bad idea for guys to work in preschools because there's a risk they'll be suspected of abuse. We heard this rumour somewhere, and I got in touch with Arbetsförmedlingen and they were like, 'What! Have our employment officers been saying things like that?' They denied it; they always do. So, we want to talk to the jobseekers directly.-Olle, Preschool Teacher Here, there was a need to create a process around what the municipality had seen as a lie, a rumour spread from a public authority. Both parties consider this as the start of a collaboration. There is an external party, and the focus is to strengthen the preschool as well as the work of the municipality; hence, the quotation can be viewed as an example of a multi-dimensional process. The target has not been clearly stipulated and none of the parties knows exactly what will result from this joint work.
One municipality has a more socially innovative grasp of the issue. This is based on an attempt to create more processes to include different groups of men in the work with preschools: One of the guys in the group had this great idea that we needed to get in contact with unaccompanied male refugees so they can get out there and do a work placement. So, I got in touch with the woman who is the head teacher's management support and said that we were interested in talking to this group, and she thought it was a good idea. Because we know that people learning Swedish might need a break from their studies sometimes, get out for a week and practise the language. And that's what we're thinking, one or two weeks.-Charlie, Preschool Teacher Here, the process is based on an idea that there are young men in the community who need contact with the labour market. The quotation shows how an idea is on the way to becoming a multi-dimensional process with collaboration with groups that are unknown to each other. There is an unclear goal with the preschool in focus, but that engages society as a whole.
In conclusion, it is possible to argue that analysis of such processes shows that these can all appear in very different ways, but the key is that the result is not completely known and several parties are involved; the preschool is in focus and the social tasks of preschools become visible. The analysis also shows that it is not always clear how situations or statements are to be analysed. This is a self-evident result, as analyses of this kind are never concise or based on statistics-there are always unknown areas. It is due to the fact that despite all the ambitions, people and organisations cannot always manage the aspects logically and with thorough reflection.
These three types of multi-dimensional processes conducted in the municipalities on the subject of men in preschools also show how the theme can be addressed in a range of different ways. The municipalities have great freedom to choose how they want to work with the topic, if at all. There is nothing specific that regulates the work of recruiting more men in preschools; however, several of the municipalities that were analysed believe that important support for the work includes demands in the education legislation governing who may be employed in preschools, together with the task of preschools of achieving gender equality.

Conclusions
This paper provides an account of knowledge concerning municipalities' actions in embedding particular types of work in order to achieve a gender balance among the staff in preschools. These types of work have been described here as initiatives, onedimensional and multi-dimensional processes of work. Initiatives, one-dimensional processes and multi-dimensional processes, characterise the work carried out, starting with initiatives leading a municipality to starting processes.
The processes described form relational agency and boundary spaces as parallel processes over time. Edwards' (2005Edwards' ( , 2011 concepts were central to the analysis done, and through this study they can further be discussed and developed. This study suggests that relational agency and boundary spaces are concepts under constant formation and negotiation in an organisation. Styhre (2014) states how actions are forming the institution, and the relational agency created is forming the institution and the understanding of it and its mandate. Therefore, it can be said that giving time to create relational agency and boundary spaces are ways of creating the municipal agency concerning issues such as men in preschool.
This study contributes to the concept of "relational agency" (Edwards, 2005) by showing how creating such an agency is a process and cannot be expected to produce results immediately appear immediately. Time is a factor in the ways of working described, since, as it is shown here, a municipal agency can be said to start when an organisation is created. A relational agency between group members creates team learning or what was stated earlier, team agency, and a relational agency can develop where boundary spaces are identified and acted on. Depending on the degree of willingness and preparedness in the various part of the organisation, a relational agency can create limited or substantial space for development.
It is through a relational agency that the municipal agency is performed, and a team operationalises often the organisation. A team that is created develops team agency, in terms of the kind of content with which to approach the question of recruiting more men, when the team members start recognising each other's expertise.
All three ways of working that have been described in the results section can take place either one at a time in a more linear way, or in parallel. The ways of working in different kind of processes to recruit more men into preschools show how differently the issue can be approached, and the types of relational agency that are created restricts the possible results. If the team created to organise the work does not find productive ways to recognise each other's expertise, a boundary space, or if the right kind of expertise is not present among the members of the team, then the relational agency, is likely to be limited.
The work of recruiting more men into preschools rests on combining preschool expertise with HR expertise and gender equality expertise. One of the conclusions that can be drawn here is that. when the work is based on the needs of the preschool and preschool organisation, it is viewed as successful by those who are running it. This may be because the mandate for change and development is already established in the preschool organisation, and HR and gender equality is viewed as a supporting expertise with specific and important knowledge. In cases where a municipality, for various reasons, does not include all three knowledge areas equally, the same experiences of direction and ambition cannot be found. This can be said to be the constitution of relational agency concerning recruiting more men to preschools.
There is a paradox in the work that municipalities need to manage and be aware of. The analysis bears witness to how men seldom want to be singled out as just "men in preschools". They want to be seen as preschool teachers or child carers. This paradox can be addressed in different ways: for example, how the measures and processes have a different focus, depending on the target group or recipients of a measure. Normalising men in preschools while simultaneously working to increase the number of men is both important and possible. The notion that men will contribute something special still remains though. Knowledge of gender, equality and the task of the preschool can change the conditions for men in their workplaces and how strongly the minority position of "being a man" is felt and experienced.
The research suggests that, over the years, the network studied has been active and has worked quite effectively, with the development of considerable knowledge around the theme "men in preschools" among the individual participants and municipalities. The levels of knowledge and awareness within the municipalities are important regarding the subject of men in preschools. In different ways, many participants refer to an increase in knowledge or awareness, among themselves, their groups or the entire organisation. This also affects the quality of what is carried out and, as a result, the accuracy, i.e. the idea that municipalities use their resources correctly. The municipal work that has been organised has become clearer and reflects a deepened awareness of the complexity around the issue of recruiting more men into preschools. As the work has been aimed at bringing men into preschools, it is possible to see a shift in it. as it generally becomes more about long-term goals, structural changes and quality assurance measures in preschools, rather than rapid changes. This does not mean that the more specific matter of men in preschools has been neglected dropped; the issue has instead been expanded. In certain cases, it has become less focused on the matter of what men do in preschools, and instead focused on recruiting more men to work in preschools.
The implications of this study are several. First, it is important to bear in mind that recruiting more men to preschools, or any early childhood education setting, is a matter of changing structures, something that never happens fast. This work requires a longterm approach with a clear goal and a local and national strategy. The lack of a national strategy is problematic since individual municipalities do not have the mandate to change these kinds of structures alone, and therefore national strategies are needed to create a direction for the work done.
Without a team or an organisation, the municipal agency is hard to accomplish. What has also become clear is that different actors at different levels in the municipality need to problematise their own level of knowledge and their own presupposed ideas of men working in preschools. It is at these workplaces men are going to work, and men need to be welcomed in the same way as any preschool teachers.
The issue of what is to be done to increase the number of men in preschools cannot be resolved based on this research; however, it is possible to state which approach appears to work.