Making Sense Across Levels in Local School Governance : Dialogue Meetings between a Superintendent and Subordinated School Leaders

Dialogues and relations between interdependent leaders working at different hierarchical levels within a given school governance system are crucial for developing shared understandings which are seen as a prerequisite for school development. Shared understandings among interdependent actors emerge from productive and dialogical sensemaking processes. The current study provides insight into how sensemaking plays out in dialogue meetings set up by a school superintendent and a team of subordinated school leaders, initiated to establish and maintain a shared interpretation community working with significant areas of pedagogy and schooling. Drawing on action research with observations, reflective conversations, and reflection notes from five central participants in the local school system, and framed within a theory of sensemaking, this issue is addressed by demonstrating how dialogue meetings strengthen the relations between a superintendent and school leadership teams. In such a context of asymmetrical power relations, the current study argues that sensemaking constitutes the pivotal activity in dialogue meetings as it fosters productive relations and bridges the gap between municipalities (as school districts) and schools. In the dialogue meetings subjected to the study, steps were taken towards shared understanding, and the involved leaders set the tone in this process by acting as democratic role models and as facilitators of creating space for reflection.


Introduction
There is today consensus among researchers and practitioners that productive relations between the chains of a given school governance system are vital for the successful adaptation of reform intentions in school districts and schools (Datnow, 2002).At the heart of this analysis lies the argument that school agents, working at different levels in the

Background
Local dialogue meetings, the subject of the current study, are also a part of a system of governing dialogues.According to Bukve (2009), this system, which is widely used in Norwegian public administration, stresses collaboration, dialogues and networking in parallel to regulations and external control, with the intention to soften the steering of lower levels.Despite this soft rhetoric in policy documents, an inherent challenge of such dialogue meetings might be to tackle the asymmetrical power relations inherent in meetings of actors residing at different hierarchical levels.In this perspective, the dialogue meetings can be interpreted in a Bildung discourse-focusing on democratic ideas, relations, and communication and a step aside from the dominant outcome-based discourse focusing on achievement, test results, and accountability for subordinated school leaders.
In Norway, the National Quality Assessment System for Education requires local education authorities to have a continuous dialogue with professionals about quality in local schools (Ministry of Education and Research, 2008, p. 55).This objective can be achieved through dialogue meetings between the local school administration and schools.At the beginning of this century, many Norwegian municipalities pursued school development through dialogue meetings (Berg, 2015, p. 14).
Although studies on superintendents and school districts and their importance in school development are growing (Nir, 2014;Moos, Nihlfors, et al., 2016), there is little research on meetings across levels in a local school setting.However, Roald's (2010) research stands out.He argues that dialogue meetings can strengthen schools' and municipalities' work on school development, especially if the meetings include several participants from the schools and the municipality, if the schools keep control over the meetings, and if regular meetings are held at the schools' premises (Roald, 2010, p. 225).Moreover, dialogue meetings can be valuable for both parties, even if they are embedded entire municipality, the head of the school leaders, and subordinated to a political board (Paulsen, 2014, p. 407). in asymmetrical power relations; the schools receive advice, support, and attention to their concerns, whereas the superintendent gains insight into what goes on in the schools and the possibility to identify and initiate improvements necessary for the schools (Engeland, Langfeldt, & Roald, 2008, p. 191).However, I have found no research with observations of dialogue meetings.Therefore, by qualitatively examining dialogue meetings, this study aims at gaining insight into important research objects which are under-investigated in prior research on school governance.

Theoretical framework
In a complex multilevel public education system, it is challenging to interpret and make sense of partly conflicting policy goals (Hooghe & Marks, 2010).Still, actors within a governance system must continuously create their interpretations of the environment, especially when unexpected things occur and there is a need for explanations and shared understandings.This process is conceived as sensemaking, a continuous process by which interacting individuals try to give meaning and make sense of their surroundings by talking about their experiences."Sensemaking involves the ongoing retrospective development of plausible images that rationalize what people are doing" (Weick, Sutcliffe, & Obstfeld, 2005, p. 409) because organisations in their origins "are vast, fragmented, and multidimensional" (Daft & Weick, 2001, p. 242).Moreover, Weick (2001, p. 4) argues that since decisions are products of sensemaking, sensemaking is a more significant concept for organisations than decision-making.In loosely coupled organisations, sensemaking is especially central: "Managers literally must wade into the ocean of events that surround the organization and actively try to make sense of them" (Daft & Weick, 2001, p. 244).Interpretations within this system are keys to decision-making, and leaders play a crucial role, as organisational interpretation results when leaders converge on their understandings (Daft & Weick, 2001, p. 243).
Sensemaking is fundamentally about communication, but it is also about action: When people are making sense, the question of what to do next immediately arises.Sensemaking both informs and constrains action (Weick et al., 2005).Individual interpretations are talked into life and adjusted as people communicate with each other.When their understandings converge, a basis for further action is formed.
Sensemaking might be used to deal with what Moos, Nihlfors, et al. (2016) see as loose couplings and broken chains in the Nordic governance systems.Establishing shared understandings-across the levels-in significant areas of pedagogy and schooling might tighten the couplings and repair the chains.Building a shared interpretation community where the participants reassess their practices makes the participants better able to understand how the system works, and by that, shared understanding emerges.Sensemaking anchors aims and purpose, just as "a range of translation and sense-making practices [are] employed by municipal managers in order to make central aims adaptable at 'street level'" (Moos, Paulsen, Johansson, & Risku, 2016, p. 287

Method
This study followed the implementation of dialogue meetings in a small-to-medium Norwegian municipality, consisting of slightly fewer than ten primary and lower secondary schools.The superintendent visited each school in the municipality and had a dialogue meeting with the school's leadership team, each half year.The meetings, concerned with school development, centered on 1) the school's status (test results, surveys, experiences); 2) the school's challenges (e.g.pedagogics, cooperation with other municipality units, leadership, economy, ICT, buildings); and 3) discussion of initiatives for improvement.
Action research has been employed to dig deep into the change process and support the practitioners in their work.Action research is not concerned with the world as it is but rather with change processes, applying a spiral of action cycles undertaken to improve practice (Herr & Anderson, 2005, p.5).The interventions consisted in helping the superintendent with planning the dialogue meetings, but most importantly in being a critical friend (Swaffield, 2008), supporting and questioning practitioners through reflection notes and reflective conversations, in an effort to improve the meetings, to promote reflective practitioners (Schön, 1983), and to understand the processes.
Data was collected for the first year and a half of the dialogue meetings.Each half year, after the dialogue meetings, five participants (the superintendent, her advisor, and three school leaders) wrote a reflection note, had a reflective conversation with the researcher, and wrote another reflection note.In sum, the main collected data consists of observations and audio recordings from 9 dialogue meetings (dm)3 , each lasting about 1.5-2 hours; 16 reflective conversations (conv), each lasting about 30-60 minutes; and 20 reflection notes (note), each about a page (see Appendix for an outline of the data collection).Unfortunately, some audio recordings were corrupted due to a technical error and could therefore not be transcribed.However, the missing data did not ruin the study since the research log to a certain extent covered what was missing, and the total amount of data still provided a foundation for analysis.
I have used qualitative data analysis software, NVivo, to organize and keep track of the data and the coding, with an easy way of maneuvering between it all.Through inductive reasoning, the initial coding identified areas such as participation, preparation, trust, aims, follow-up, openness, experience of the dialogue and system understanding.When sensemaking emerged as a central theme, data were categorized into different aspects related to sensemaking, ending up with shared understanding, openness, critical questioning, visions, participation and system understanding (reflected in the next part).In the writing process, I have gone back and forth between transcripts and drafts, re-categorizing and searching for keywords while describing, analysing, and discussing.It has been an eclectic process, which is normal, according to Creswell (2012, p. 238), when analysing qualitative data.
The well-known challenge in action research of the double role of being both a researcher and a practitioner (Levin, 2012), has been handled by separating the two in time.With about three years gone since my role as a critical friend ended, I would contend that sufficient analytic distance has been secured.I have tried to keep an open mind and to be aware of my presuppositions and biases.Still, it is unlikely that I am free of bias since I have invested effort in the project and am therefore more likely to view it as a success (Coe, 2009, p. 366).The participants also have their presuppositions, so in interpreting their perceptions one should note that the participants most likely suffer from positive bias.Nevertheless, I have tried my best to give an accurate account of the dialogue meetings and the participants.I also believe the results deserve attention, even if they might be somewhat inflated due to positive bias.
As recommended by Levin (2012, p. 144), I wrote my presuppositions in a research log before the dialogue meetings started.I expected that it would be easier to anchor aims and purpose when several of the team leaders joined a meeting with their school leader and the superintendent.As we will see, the findings show that this expectation was warranted.Additionally, I expected organisational learning to be central, much more than what the findings indicate.However, sensemaking was not in my mind until long after the data collection ended and discussions with supervisors on preliminary drafts and reading of theories brought it to the forefront.Steps to secure validity have been taken through triangulation and member checking (Creswell, 2012, p. 259) and by opening up the study for inspection, making all components transparent (Eikeland, 2006, p. 231).Triangulation was secured by validating findings from different individuals and across different methods of data collection (observations, conversations, and written notes).Member checking was done by letting the participants read, check and comment on the accuracy of the study.

Building a shared understanding
The participants were asked about their view on the process in the reflection notes.After the first dialogue meeting, school leader B wrote about the lack of coordination, stating, "There's been created some confusion about the logistics of this process" (note 1).School leader C agreed, saying, "I think there are too many plans, too many different concepts, and too many processes going on at the same time" (note 1).However, already after the next dialogue meeting, their impression was quite the opposite.On the same question school leader B now wrote, "The process is good" (note 2), and school leader C wrote, "This is neat and straightforward.No problems" (note 2).This change of mind was also confirmed in the reflective conversations.School leader A, however, thought everything was clear all along the process.My impression is that school leader B and C got a better grasp of how everything fitted together as the dialogue meetings proceeded.
In the first reflective conversation, the superintendent said that the purpose was to "try to understand the schools, first and foremost" (conv 1).She wanted to make sense of what was going on in the schools.She continued: "It's something about understanding it well enough before one comes with suggestions, showing respect for those who are in the school on a daily basis" (conv 1).The superintendent tried to build a shared understanding, bridging the gap between the municipality and the school, cutting to the core of sensemaking by asking the two first questions of the process: "What's going on here?"and then orienting towards the action component, "What do I do next?"(Weick et al., 2005, p. 412).Later in the process, the superintendent confirmed that she had a better understanding of the schools (note 2).
In the third half year, the superintendent said, "The dialogues are increasingly improved.They have a development which makes us 'speak the same language' more" (conv 3), implying that it is easier to understand each other as they all try to make sense of how to work on school development.The school leaders supported this shared understanding.School leader A wrote, "I think there's an understanding of each other's viewpoints" (note 1), and school leader C wrote, "I feel we're met with a good deal of understanding from the employer's side" (note 3).When the participants try to comprehend each other's viewpoints, they build a shared understanding.This dialogical approach to building shared understandings finds resonance in research on superintendents: One gets the impression that superintendents have clear and high intentions to establish relations with school leaders with a high content of sense-making, and they aim to do this through sparring, exchange of experiences, discussions of strategies, coordination and collaboration.(Moos, Paulsen, et al., 2016, p. 303) School leader B emphasized having broad participation, "It's important for me to have several people from the school; I think it ensures that the dialogue is anchored too.If not I'd have to be very conscious when taking the dialogue onwards" (conv 3).A shared understanding among many of the school's participants makes it easier to convert points from the dialogue into the school's daily practice.The superintendent's advisor agreed on this point, stating, "I think it's easier for the school leader when more people are involved, whether it's the leadership team, the teacher representatives, or others when one wants to create a shared understanding at the school" (conv 4).The key words here are to create a shared understanding at the school, which is more easily attained through broad participation in the dialogue meetings.
In the third half year, a shared understanding was seemingly created in school B when they talked about their high scores on gender equality.School leader B first said he did not know why and that he was surprised at the scores.However, after several of the team leaders aired their opinions, the high scores made sense to them all.The superintendent summed it up; they had, among others, practical approaches to learning, concretization material, and several practical elective subjects (dm 3B).Now they had a shared understanding of the causes for the gender equality in the school.Additionally, they now had ideas they could share with other colleagues inside the school, but also in other schools.
In school C, a team leader explained how they planned to follow up on pupils' test results through a culture of sharing: We've made a plan for the fifth-grade teachers.They go through their tests, one and one, and then all the fifth-grade teachers try to analyse together with me, where the challenges are.Then the teachers convey this to the team leader and the teachers in the third and fourth grade, so they get some input on what they should focus more on.(dm 3C) Several participants in school B also wanted more internal sharing.A team leader wanted dialogues with, and feedback from, teachers at higher grades; a teacher asked for sharing between teachers at the same grade; and another teacher envisaged that observation in the classroom could be beneficial (dm 1B).It is evident that these participants desired a better culture of sharing and learning.Joint reflection and sharing across and within grades is a step towards learning about teaching and making sense of one's own and others' experiences.Related to the dialogue meetings, the participants received ideas on how to build cultures of sharing and learning.
At the end of the meetings, some of the participants expressed their impressions of the meetings.A team leader said, "I think it's very inspiring, I get the urge to go on, to really get a hold of it and work on the things we've set up here," and another team leader immediately consented, "Yes, I agree.That's all I say" (dm 3B).At another school, a teacher who was openly critical towards the municipality also commended the meetings, "It's useful, talking, or conducting a dialogue where we, in a way, can say things" (dm 1A).It seemed like they all found it useful to talk to each other.When the superintendent explicitly asked for the participants' impressions, a few of them joined in and gave their impressions.In so doing, they contributed to building a shared understanding, and they gave the superintendent and the school leaders the opportunity to check the participants' understanding.If every participant had been asked for their impressions, systematically, it would probably strengthen their shared understanding.

Securing openness
The dialogue meetings opened up communication between participants at different levels.School leader C wrote, "I feel that the dialogue is very open and honest, and we all say what we think" (note 2).The superintendent concurred, "I believe the impression is that it's been a harmless forum" (conv 4).The observations support this.The agenda was flexible, and the participants were free to speak their minds.Questions were asked, disagreements and spontaneous discussions arose.The participants discussed challenges with different views and talked openly about the negative aspects of their school.One example is when a team leader talked about students who were afraid of other students: "If it were my child, I wouldn't tolerate it at all.We see it every day, but we don't have the equipment or the tools to do anything with it" (dm 1C).This openness presupposes trust among the participants-trust that allows challenges to be dealt with before they become too complicated (Louis, Mayrowetz, Smiley, & Murphy, 2009, p. 161).The atmosphere was in general relaxed and good-humoured, illustrated by laughter from the participants on several occasions.Many of the participants spoke-not only the superintendent and the school leaders.In short, there seemed to be a culture for open and trust-based discussions.
Looking critically at the degree of openness, some participants hesitated to say what was on their mind.One team leader explicitly expressed this, "To be honest, I find it very difficult to sit here and talk about it, because much of it's about leadership, and it ... yes, I find it difficult" (dm 1A).This utterance is a clear sign that not all participants felt they could say what was on their mind, especially not when an internal dispute was aired, as in this case.However, the asymmetrical power relations-between the superintendent and the other participants, but also between the school leaders and the team leaderscould play a part in restraining the discussions.Even though a few of the participants hardly spoke, it is hard to know whether this was caused by the asymmetrical power relations.The leaders could have encouraged the expression of conflicting points of view and, more often, involved all the participants.Still, my general impression is that the dialogue meetings were a safe space in which teachers and leaders could question each other and discuss ideas and challenges-in other words, a supportive group-climate, characterized by psychological safety (Paulsen & Henriksen, 2017, p. 80).

Critical questioning
School leader C highlighted the value of the dialogue meetings, "It's necessary to discuss results with others and not only with ourselves.It makes us more critical, and it calls for more explanations and analysis than for those who 'own' the results" (note 1).In other words, the dialogue meetings made the school investigate more.The superintendent continually expressed a wish "to be more critical, to drill a bit deeper into the matter" (conv 1) and to have "more critical and prepared questions to each school" (note 2).Research on the relationship between superintendents and school leaders underscores sparring as valuable (Paulsen, Nihlfors, Brinkkjaer, & Risku, 2016, p. 225), thus critical questioning is essential.
Leaders can promote reflection and critical questioning towards content, process, structure, aims, and purpose.From a learning perspective, it is especially vital to question, challenge, and change operating norms (Morgan, 1998, p. 83).However, the critical questions in the dialogue meetings mainly touched upon content and structure, whereas other questions were lacking.Still, there were some tendencies.The superintendent had a question that stirred the seemingly univocal view that the dialogue meetings were good: "Would it be just as positive if you'd come in and taken a cup of coffee and asked, 'How's it going?'"(conv 4).With this question, the premises behind the dialogue meetings were questioned, and a discussion about purpose could have followed.However, the question was not pursued further since it came up in the last reflective conversation.In a dialogue meeting, school leader C queried, "What distinguishes a teacher from one who's not a teacher?" (dm 3C).However, this question about the very foundation of teaching did not result in a discussion because school leader C already had a ready-made answer related to the importance of planning and teachers' ability to make plans, act and evaluate.To sum up, critical questioning lacked in the dialogue meetings under study.However, critical questions can open up discussions and move the organisation towards organisational learning (Morgan, 1998, p. 82).

Leaders paving the way
In the last reflective conversation, the superintendent began by stating the purpose of the dialogue meetings: "It becomes more and more evident, the importance of building a community and dialogue meetings should be used to assess central elements in that community, educational platform, leadership" (conv 4).The superintendent wanted to build a community of cooperation on local school issues with dialogue meetings as a central part.Researchers claim that "if the goal is to alter traditional patterns of responsibility and leadership in the school then groups are the most effective unit of change" (Louis et al., 2009, p. 159).The school leadership team can thus be an effective unit of change.Together with the superintendent, they can set the agenda for local school development, lead the creation of shared understanding, and build images of themselves as leaders of change.Daft and Weick's idea of organisational interpretation, "interpretation by a relatively small group at the top of the organizational hierarchy" (2001, p. 243), fits with the participants in the dialogue meetings as they are a relatively small group at the top of the local school organisation.
In dialogue meetings, the superintendent and the school leaders are seen and experienced by the other participants.From a cultural perspective, "You are what you are seen and experienced as being, not what you think you are or what your job title or job description says you are" (Morgan 1998, p. 142).Hence, in the dialogue meetings, superintendents and school leaders communicate and demonstrate the organisation's visions and values through talking and acting.Leaders can be influential role models in creating a culture of sharing and building shared understanding.Throughout the dialogue meetings, the superintendent pursued this by constantly expressing the goal of improving pupils' learning with enthusiasm: Together we'll make a great show, the growing-up show, and then it's important that we have an anchor and that we contribute in that direction, and maybe it makes it a bit easier at the schools too, if we're doing more of the same when we meet, that's some of the intentions.(dm 1C) Here the superintendent demonstrated high ambitions, stressed the importance to secure plans from top to bottom within the organisation, to cooperate and to do "more of the same" across the schools for example by having more similar methods and common standards.In other words, the superintendent tried to build a sense of cooperation, shared responsibility, and a shared vision.
The school leaders were also aware of the importance of a joint effort, and of their role as models of ideas and visions.School leader B said, "I think it's been important to be transparent, so everyone understands that the boss wants the best for the teachers.I think that's the first priority" (conv 4).With many participants from school B, they all could then see and experience their leader modelling key ideas of democracy by being open and honest in the reflections, by showing that they are in this together, by securing involvement and equality in the process, and by pointing towards a shared vision for the school and the municipality.Leaders can motivate the participants and shape and guide organisational action as "organizations end up being what they think and say, as their ideas and visions realize themselves" (Morgan, 1998, p. 139).School leader C insisted that school leaders have to follow directions from their superiors, "if the leadership level doesn't manage to have a culture of leadership, so I listen to you [the superintendent] when you say this is how we do it.If not all leaders do that, we're done, so we have to start there" (dm 1C).In other words, school leaders must cooperate and coordinate their work with their superintendent before they can expect the teachers to listen to them.In this respect, dialogue meetings can be a step in the right direction towards more coordination at the leadership level.

Democratic leadership intentions
As mentioned, there was a clear asymmetrical power structure in the meetings.The superintendent, having formal power, was in charge of and led the dialogue meetings, set the agenda together with her advisor, and the advisor wrote the meeting minutes.Thus one can say the superintendent had control over what Morgan (1998, pp. 165-166) distinguishes as the premises, processes, issues, and objectives of decisions.This sort of control supports Paulsen's (2014, p. 414) suggestion that Norwegian superintendents act as gatekeepers who decide which tasks should be brought to the table and which should be downplayed.
Still, the superintendent did not control everything.The schools decided in advance a few of the issues on the agenda, and the participants were free to raise any topic they wanted in the meetings.School leader A underlined the significance of equality, "It's important that we're all on the same level from a power perspective, a democratic perspective" (conv 3), suggesting that this equality existed in the meetings.An open and honest dialogue where everybody has a chance to speak results in a more consensus-oriented meeting in line with democratic forms of local school governance.Dialogue meetings can then be an intentional step towards leveling out the asymmetrical power relations.
A democratic leader resolves tensions and conflicts by asking, "How shall we do it?"(Morgan, 1998, p. 152), involving more people in sensemaking processes on what is going on and what to do next (Weick et al., 2005, p. 412), including all voices, not only those who speak unasked.Moreover, a democratic leader spends "time listening, summarizing, integrating, and guiding what is being said, making key interventions and summoning images, ideas, and values that help those involved to make sense of the situation with which they are dealing" (Morgan, 1998, p. 171).This democratic leadership style matches the dialogical aspect of the dialogue meetings.The data also indicates that the superintendent was perceived as a democratic leader as several of the participants complimented her leadership style.A team leader said, "I think it's very interesting and the dialogues are good, and you're skilled at leading these meetings" (dm 3B).With knowledge of how the system works, the superintendent often explained things to the school's participants but also listened to their concerns.Paulsen et al. (2016, p. 212) argue that having enough meeting points is crucial for avoiding a broken governance chain in the local school hierarchy.In this regard, dialogue meetings can help to keep the local governance chain intact.Agents from different levels in the organization meet to cooperate on joint challenges, trying to understand and find their way through the complexity.In summing up the whole process, school leader B touched upon how the dialogue meetings helped bridge the gap between the municipality and the school:

The chain of governance
From the first dialogue meeting, I felt that we were told to pinpoint areas for development.Perhaps these areas were more characterized by our own goals, and not related to the central goals of the municipality.So there were more like two worlds earlier than it's now.I now feel that our goals are more like measures to achieve the municipality's central goals.(conv 4) School leader B suggested that earlier the chain was broken, but in the dialogue meetings, the school's goals were related to the municipality's goals, indicating an intact chain in the local school governance system.
In the last reflective conversation, the advisor reinforced this view, stating, "The dialogue meeting in itself has helped create closer cooperation, a more unified understanding between the municipality administration and the schools" (conv 4).It seems like sensemaking as an ongoing process in the dialogue meetings is central in bridging the gap between the municipality and the schools, suggesting a collaborative climate between superintendents and their school leaders in line with prior research (Paulsen & Henriksen, 2017).

Discussion
Creating space for reflection Dialogue meetings can facilitate openness and honesty in discussions, and by that, be a place for sensemaking and learning.Both, which are dependent on reflection, "the cognitive activity of attempting to make sense of experiences" (Seibert & Daudelin, 1999, p. xi).The dialogue meetings emerge as spaces where the participants try to make sense of the complexity of test scores, surveys, and experiences.They get a better understanding of their own and their colleagues' practices but also a better understanding of the municipal school system.
Sensemaking initiates understanding and learning at an individual level, but more decisively, at a collective level.Sharing information by talking and reflecting on experiences allows the participants to converge on interpretations (Daft & Weick, 2001, p. 243), making their understanding more alike.The participants are "creating networks of interaction that can self-organize and be shaped and driven by the intelligence of everyone involved" (Morgan, 1997, p. 116).Openness, trust, and psychological safety are essential to fruitful reflection in these networks.
To trace school development, it is crucial to follow up key points from previous meetings.The participants will also find the reflections more worthwhile when the meetings are followed-up.Additionally, the reflections increase in value if all participants are included, for example by letting everybody reflect at the end of each dialogue meeting.However, one must dare to ask critical questions, particularly towards aim and purpose of the dialogue meetings.Asymmetric power relations might deter subordinates to be critical, so it is necessary to clarify that criticism is welcome.For example, leaders can set time aside for reflection in advance, to prepare the participants, and leaders can be role models by asking critical questions.On the other hand, critical questioning could expose underlying conflicts in the group.However, competent conflict management can encourage reflection and yield positive outcomes for example by challenging what has been taken for granted (Morgan, 1998, p. 175).In sum, these points could tie efforts closer together, increase participants' involvement in the process, enhance shared understanding, and be a weighty contribution to the sensemaking process.

Towards democratic ideas
Even if the superintendent set the premises and steered the dialogue, the findings point towards a reduction of power play in the dialogue meetings.The participants were on a more even level in the meetings than what the asymmetrical relations suggest.After all, the schools contributed with a few points on the agenda and the participants were free to take up any issue they wanted.Moreover, the other participants also praised the meetings.According to the five key participants, dialogue meetings are good sites for becoming familiar with each other, for support, evaluation, and setting aims for further development.
Dialogue meetings can be a central space for developing a culture of equality characterized by symmetric relations, trust, openness, and transparency.With a democratic leadership style, the participants reconcile their differences through consultation and negotiation, and they converge on a shared understanding while trying to make sense of it all.Dialogue meetings are a contrast to traditional hierarchical structures, a step away from neoliberal ideas with an emphasis on control and accountability mechanisms and a step towards a Bildung discourse.This move towards a culture of equality finds resonance in research claiming that superintendents enter into a dialogue between equals when they try to make sense of local school issues with school leaders (Moos, Paulsen, et al., 2016, p. 303).
However, dialogue meetings might also go in the other direction, towards neoliberal ideas, strengthening the hierarchical structure, increasing control and the use of instructions, since words and actions-not the form-constitute the meetings' essence.Dialogue meetings alone are not enough.One must safeguard the democratic ideas to achieve an open, critical and constructive dialogue.
Team leaders have a critical role in this sense, as they are positioned at the interface between the school leader and the teachers, being able to influence in both directions.Team leaders' active involvement builds trust in dialogue meetings since their positioning in the middle facilitates and nourishes a culture of trust (Edwards-Groves, Grootenboer, & Ronnerman, 2016, p. 384).Team leaders are therefore essential to include in dialogue meetings when striving for democratic ideas of leadership and citizenship.
The data also suggest that the dialogue meetings provide the superintendent with a better understanding of the schools and the school leaders improve their understanding of the system.The result is a greater understanding that converges towards a shared meaning and understanding, bridging the gap between administrators and practitioners, that is, the municipality and the schools.Extending dialogue meetings upwards, so local politicians have a say, would make dialogue meetings democratic in its original meaning and reduce the trend of increased administrative control of local school-policy (Engeland et al., 2008, p. 195).

Conclusion: The pivotal role of sensemaking
School development with consistent changes in teachers' practice is beyond the scope of this study.Still, I believe the achievements should be considered substantial.The study indicates that dialogue meetings can yield productive relations between levels in the local school hierarchy, bridging the gap between superintendents and school leadership teams.With sensemaking in a central role, dialogue meetings can be a step towards establishing shared understandings with joint reflections and discussions and a step towards a more democratic process with broader involvement.Leaders are role models and facilitators of creating space for reflection.They must also safeguard the keys to successful dialogue meetings: trust, openness, broad participation, systematic follow-ups, critical questioning, and democratic processes.Sensemaking has a pivotal role, as the participants find meaning in building a shared understanding of the dialogue meetings, in the dialogue meetings, about the schools, the system, and about themselves as leaders of change.Sensemaking makes the participants more willing to partake in the joint endeavor of improving schools.They feel more united as the chain of governance is made visible and meaningful.