Leading for higher grades—balancing school leadership on the fine line between accountability and professional autonomy

ABSTRACT

Caught in the crossfire of expectations of improving students’ learning outcomes while simultaneously safeguarding students’ well-being, school principals must balance accountability and professional autonomy. This article presents findings from a small case study that examined the relationship between the supervisor and four school principals in a Swedish municipality. Drawing on an institutional perspective, the purpose of the study was to examine how the superintendent manages the principals through the quality assessment system through regulative, normative, and cognitive elements and how the principals relate to these institutional aspects. The empirical data was collected by interviews. Scott’s pillars of institutional order were applied to analyze what regulations, normative expectations, and cognitive elements that could be identified in the empirical material. Bridging and buffering were used as analytical tools to analyze the strategies used by the principals to relate to these institutional aspects. The findings indicate clear and well-implemented regulative routines, but also monitoring and a range of sanctions where the principals do not meet expectations. There are examples of normative elements emphasized by the superintendent, as well as cognitive elements, which provide the infrastructure of the organization. To handle these institutional aspects, school principals use adaptive strategies, such as bridging and buffering.

KEYWORDS:

• Principals
• superintendent
• quality
• school improvement
• boundary practice
• loose coupling

Introduction

Based on findings from a small Swedish case study, this paper examines the relationship between a supervisor and a sample of four school principals in the context of school improvement. The following key terms are used to characterize the practice and processes involved, namely ‘boundary practice’ (used to define the quality assessment system as a routine that sustains connections between the superintendent’s and the principals’ communities of practice); ‘loose coupling’ (whereby school principals had a relationship with the superintendent but still maintained their own professional autonomy); ‘regulative’, ‘normative’, and ‘cognitive’ elements (providing norms, rules, and definitions for school leaders’ actions); ‘bridging’ and ‘buffering’ (which characterize the strategies used by principals in an effort of both ameliorate and reduce the influence of the supervisor on their schools). From an international perspective, the Swedish case is of special interest because the introduction of the quality concept in education, followed by educational reforms, within a very short period has changed the perception of education. These changes are reflected in the language used, a recontextualization of the content of education and changed relations between local autonomy and national control (Bergh, 2015). It is also significant how global education policies have been translated into Swedish national policies and frameworks for action and how influential policy actors such as the EU and OECD have set the agenda for quality work in Swedish schools (Grek & Lawn, 2009).

In the field of education, the global policy trend is characterized by mounting pressure to improve students’ learning outcomes at different levels in the school system. Increased demands in recent decades for efficiency and performance have led to a culture of measurement, with performance-based management becoming the ideal (Møller, 2009). This has been accompanied by a standard-based accountability movement (Au, 2011; Chatterji, 2002; Kim, 2018; Sivesind & Karseth, 2019; Waldow, 2014). The shift in international policy discourse toward an increased focus on measurable outcomes can be referred to as ‘managerial accountability’ (Møller, 2009), which further can be translated into superintendents holding school principals accountable for student results. As part of the accountability movement, data-informed decision-making has become a relatively new aspect of principal leadership (Shen et al., 2016). High expectations are placed on school principals to focus on student performance and make use of performance data to improve teaching and learning. This implies new professional demands for school principals. Although data as a tool for school improvement is not something new within the field of education, the focus on accountability for student performance has increased in recent decades (Shen & Cooley, 2008).

Data gathering has become an increasingly important part of quality management in education. In this ‘data-managerial-oriented’ approach, there is a strong focus on student outcomes, quality assurance, accountability, and quality control, here with implications being that superintendents, as well as principals, tend to use data of learning rather than data for learning (Prøitz et al., 2019; Shen et al., 2017). Against the backdrop of this data-managerial-oriented approach as the ideal, we have explored the quality assessment system as a boundary practice (Wenger, 1998), with the superintendent and school principals being its key actors. With routines that aim to sustain connections and tighten the couplings (Orton & Weick, 1990) between the superintendent’s and the principals’ communities of practice, our hypothesis is that the LEA’s quality assessment system as a boundary practice will produce powerful normative and discursive agenda-setting tools. As a response to these regulative, normative, and cognitive aspects (Scott, 2008), we assume that the school principals use strategies to comply with the quality assessment system, as well as buffering and gatekeeping strategies (DiPaola & Tschannen‐Moran, 2005). Drawing on an institutional perspective, the aim of the current study is to examine how the superintendent manages the principals through the LEA’s quality assessment system and what consequences are imposed for not reaching standards. Two questions guided the present study:

• What regulations, normative expectations and shared conceptions can be identified between the superintendent and principals in the boundary practice of the quality assessment system?

• How do the principals relate to these regulations, normative expectations, and shared conceptions?

In the first part of the article, the Swedish context of the study is outlined. Thereafter, the theoretical and methodological framework is described, followed by a presentation of the method and empirical data. The findings of the study are then presented followed by the discussion and conclusion.

Contextualization

Although the focus of this study is on quality work in the Swedish context, its contribution is of international interest since it provides in-depth insights into how global education policies have been translated into a particular local context. With a stronger emphasis on the quality of education emerging in recent decades, the focus on quality assessment and evaluation in many countries has strengthened. For example, with the recent Swedish Education Act (SFS 2010:800), the requirements for systematic quality work in education have been explicitly formulated and are now at the top of the Swedish policy agenda. The Swedish Education Act stipulates that the superintendent, as a representative of the LEA – as well as school principals – must organize and perform local quality work, hence ensuring school improvement. The Education Act (SFS 2010:800) further stipulates that there must be a documented quality system at both the LEA’s level and at each school (Ch. 4 §§ 3, 4). The superintendent functions as the LEA’s chief executive officer, who is responsible to the LEA for the implementation of the quality work. In the Swedish school system, systematic quality work should be conducted at different hierarchical levels, much like a ‘delivery chain’ (Barber, 2007). At the district level, the superintendent has the responsibility of ensuring systematic and continuous follow-up, planning and improvement of school outcomes. There should be a basic structure that has clear routines for the systematic quality work, for the superintendent and for the school principals to follow. In turn, each school must refine the structure and adapt its own quality work so that it can be adjusted to local needs and conditions (National Agency for Education [NAE], 2015). However, although the national regulations are the same for all superintendents and school principals, there are a variety of ways that local school actors actually work in their schools and districts, mostly because of contextual factors (Ball et al., 2011; Supovitz & Weinbaum, 2008). As a result of increased school autonomy, educational leaders experience high levels of agency and discretion in defining their own priorities and duties (Shen et al., 2016).

The global policy trend, characterized by an increased focus on holding school leaders on different levels in the school system accountable for students’ learning outcomes, implies that school principals must find ways to balance performance and accountability with professional autonomy. Becoming more accountable for performance outcomes places school principals in the crossfire of expectations. Research has found that Swedish principals are under pressure to report a range of statistics to municipal officials, but they often perceive this reporting as not pedagogically motivated, sometimes feeling irritation at the interference these statistics causes with their work and responsibilities (Hult et al., 2016). School principals still ‘play along with the system’, complying with the requirements of their Local Educational Authority’s (LEA’s) quality assessment systems. However, at the same time, principals are bending the rules and acting as gatekeepers (Adolfsson & Alvunger, 2020). Placed in the crossfire of external and internal demands, school principals can be described as having ‘a double mission’: both governing and protecting their teachers and teaching (Hult et al., 2016).

Theoretical and methodological framework

This study draws on an institutional perspective. From new institutional theory, we use the concept ‘loose coupling’ (Orton & Weick, 1990) to examine the relations of both responsiveness and distinctiveness between the local educational administrative level (superintendent) and school level (school principals) within the context of quality work in a Swedish municipality. Our focus has been specifically on the quality assessment system, set up by the superintendent to maintain and strengthen local school governance. Following Spillane and Burch (2005) it can be argued that the institutional perspective risks to lose sight of the local institutional context and tends to emphasize institutional determinism over practice and agency at the local school level. As a complement and yet faithful to the institutional tradition, we agree with Spillane and Burch (2005) that the concepts ‘communities of practice’ and ‘boundary practice’ (Wenger, 1998) from socio-cultural theory are helpful in this context. By conceptualizing the LEA’s quality assessment system as a boundary practice follows that there are routines that sustain the connections between the superintendent’s and the principals’ communities of practice. As a boundary practice, the quality assessment system also provides an ongoing forum for mutual engagement in various activities. As coupling mechanisms, the routines of the quality assessment system provide rules, norms and definitions that give meaning and direction for action. Based on this understanding, Scott’s (2008, 2014) three pillars of institutional order have been used to analyze the regulations, normative expectations and conceptions that could be identified within the boundary practice of the quality assessment system. The concepts ‘bridging’ and ‘buffering’ (DiPaola & Tschannen‐Moran, 2005) have been used as analytical tools to examine how the principals relate to the institutional order provided by the quality assessment system.

School organizations as loosely coupled systems

School organizations are usually described as ‘loosely coupled systems’ (Orton & Weick, 1990; Weick, 1976), and there are various ideas about how to deal with this looseness and the advantages and disadvantages of loose coupling (Shen et al., 2017). The proponents of managerialism argue that loose coupling is a problem and that finding the mechanisms to tighten the educational system from the top-down, and establishing order and accountability, is the key to school improvement (Morley & Rassool, 2000; Shen et al., 2017). On the other hand, there are those who argue that it is possible to take advantage of loose coupling. By doing so, they promote a bottom-up approach through capacity building and strengthening professionalization (Shen et al., 2017). The description of schools as loosely coupled systems is most often based on a simplified and unidimensional approach, ‘portrayed as the endpoint of a scale that extends from tightly coupled to loosely coupled’ (Orton & Weick, 1990, p. 205). In this description, tightly coupled systems are represented ‘as having responsive components that do not act independently, whereas loosely coupled systems are portrayed as having independent components that do not act responsively’ (Orton & Weick, 1990, p. 205). Orton and Weick (1990) encourage researchers to avoid simplification and argue for a dialectical interpretation of school organizations as loosely coupled systems, emphasizing both their distinctiveness and responsiveness, juxtaposing these contradictory forces simultaneously. This can be done by studying organizational structures ‘as something that organizations do, rather than merely as something they have’ (Orton & Weick, 1990, p. 218) or by using the concept of loose coupling for detailed and dynamic explanations, rather than for flat and static descriptions.

The quality assessment system as a boundary practice

Within a school organization, there are different ‘communities of practice’ that influence each other through connections such as ‘boundary objects’ (Star & Griesemer, 1989; Star, 2010; Wenger, 1998). A community of practice is a group of individuals who share a common area of interest and engage on an ongoing basis in a common endeavor. As an example, school principals can participate in multiple communities of practice at once. They are members of the district’s administrative leadership team, which includes the superintendent and other school principals, as well as their own leadership team at their local schools. While a community of practice creates boundaries, it also develops ways to create and maintain connections to the external environment and other communities of practice. Connections as boundary objects (e.g. documents, data, concepts, and other artifacts) and boundary activities (e.g. quality dialogs and school improvement activities) make it possible for different communities of practice to influence each other. A practice itself can also become a connection, that is, a ‘boundary practice’ (Wenger, 1998), with routines that sustain this connection. In the current study, the boundary practice of the LEA’s quality assessment system is of particular interest. Between the superintendent’s and principals’ communities of practice, the boundary practice of the LEA’s quality assessment system provides an ongoing forum for mutual engagement in various activities, such as quality dialogs. Boundary objects and boundary practices go hand-in-hand. Boundary objects connect communities of practice, function as practical guidance, but also reflect norms and meaning systems (e.g. accountability) originating outside the school (Spillane & Burch, 2005). When it comes to coupling mechanisms, the organizational routines of the quality assessment system provide or contribute to an institutionalized social order (Scott, 2008).

Three pillars of institutional order

From an institutional perspective, institutional sectors provide norms, rules, and definitions that constrain and enable action (DiMaggio & Powell, 1991). These tacit schemata define appropriate structures and give meaning and order to action (Scott, 2014). As institutions, school organizations are made up of different combinations of ‘regulative’, ‘normative’, and ‘cognitive’ elements, varying among one another and over time (Scott, 2008, 2014). These aspects will both constrain and enable school leaders’ actions. Regulations emphasize the rules, sanctions, monitoring and so forth. The normative aspect concerns the norms, expectations and attitudes that are prioritized and rewarded, while the cognitive element emphasizes the shared conceptions that provide the infrastructure on which school organizations’ beliefs, norms and rules rest and through which meaning is made. For principals, an important assignment is ‘to ascertain what elements are at play in a given context and the extent to which they work to reinforce or undercut one another’ (Scott, 2008, p. 429).

Bridging and buffering as adaptive strategies

Previous research has shown that school principals try hard to cope with different and sometimes conflicting regulations and normative expectations from the LEA and their underlying premises and assumptions, hence balancing between defiance and compliance (Adolfsson & Alvunger, 2020). Important strategies in this struggle are gatekeeping and autonomy. Acting as ‘gatekeepers’ makes it possible for school principals to manage complex problems by, on a day-to-day basis, making decisions about resource allocation and staffing and finding solutions to various problems (Hallinger, 2011). Møller (2009) describes how school principals ‘struggle with the tensions of managerial demands from the outside and their own standards for acting as professional educational leaders’ (p. 42). To handle these demands and expectations, school principals use adaptive strategies, such as ‘buffering’ and ‘bridging’ (Addi-Raccah, 2015; DiPaola & Tschannen‐Moran, 2005). Buffering can be considered an adaptive strategy to reduce environmental influence as much as possible to protect the core business of schools, while bridging is a cooperative strategy to increase the interdependence of the school and environment. For example, principals strive to build and maintain good relations with the superintendent because principals sometimes need assistance and additional resources for managing their schools (Addi-Raccah, 2015). With good relations, principals can then buffer unwanted interventions from the superintendent/LEA and become more efficient in their work (Seashore Louis & Robinson, 2012). Through spanning boundaries (Wenger, 1998), a principal can serve as a type of interface between the local school and LEA/superintendent. Being engaged in internal leadership activities at the school, on the one hand, and external activities aimed at managing the school environment to acquire resources, on the other hand, principals can better facilitate school improvement (Benoliel & Schechter, 2017; Benoliel, 2017; DiPaola & Tschannen‐Moran, 2005).

Method and data

The empirical data underlying the current case study (Yin, 2018) was collected in a small town/urban area (Swedish Association of Local Authorities and Regions, 2016) in southern Sweden. The study is part of a research project which aims to increase and deepen the knowledge of how actors at different levels in the school system can support students’ learning outcomes. The superintendent and four principals (two women and two men) from public comprehensive schools were interviewed. All informants had worked for at least five years in their profession and were selected by purposive sampling (Etikan et al., 2016). The interviews were individual, and semi structured and based on open-ended questions (Cohen et al., 2018; Gillham, 2005). The same questions were asked to all respondents and were based on a questionnaire that had been sent out to the respondents before the interview. The interviews were divided into the following main themes: background questions, questions about pedagogical leadership and questions about cooperation between different levels in the governing chain, with subqueries to each heading. Exploratory questions were asked when the interviewer considered that there was more to tell at a certain point during the interview (Gillham, 2005). Finally, the respondents could make concluding remarks about subjects they wanted to address. The interviews were recorded and then transcribed. For a better understanding of the context, local policy documents and guidelines were collected and studied.

Ethical decisions have been made, and the interviews were carried out in accordance with good research ethics (Swedish Research Council, 2017). Before the interviews, all respondents were informed about the study. Through the opportunity to decide whether they would continue participating at any time during the study, the requirement for consent was met. The requirement for confidentiality was ensured by deidentifying those who participated.

In the analysis process, the first step was to listen to the recorded interviews and then briefly read the transcribed texts. The texts were then reread, this time in more depth before being coded and categorized. To answer the first research question, Scott’s (2008, 2014) pillars of institutional order were used as analytic tools. The interviews were categorized into three themes: regulations, normative expectations, and shared conceptions. To answer the second research question, the interviews were categorized into two themes: bridging and buffering (DiPaola & Tschannen‐Moran, 2005).

Findings

In the first part of this section, the results concerning the first research question are presented, that is, what regulations, normative expectations, and dominant underlying shared conceptions we identified between the superintendent and principals in the boundary practice of the quality assessment system. In the next part, the results regarding the second question are outlined, including how the principals relate to these regulations, normative expectations, and shared conceptions by the usage of adaptive strategies such as buffering and bridging.

Regulative, normative, and cognitive elements within the boundary practice of the quality assessment system

Regulative routines

The boundary practice of the quality assessment system is characterized by clearly and directly demonstrative manuals and instructions. Based on national legal instructions that are explicitly defined in the Swedish Education Act and guidelines from the NAE, the LEA has designed and implemented a local model for systematic quality work. There is also a calendar illustrating all the systematic quality activities that should occur throughout the year. For each of the activities, there are detailed instructions. In the interview, the superintendent emphasizes the importance of these regulations and routines:

I enjoy or like to be in a well-oiled machinery, which is an organization where I do not really need to chase anything to reach the optimal organization. I like or seek to have rules, having process descriptions, having routines, that is, clear rules of the game. (Superintendent [S])

The principals describe how the school organization is very hierarchical and that the governing chain is clear; they also discuss the follow-up system:

Interviewer (I): It seems to be very clear which regulations apply; there does not seem to be anyone who hesitates.

Principal 4 (P4): No, I do not think so. There is no doubt about it, and we follow it; in that way, we are very good at following the law, I must say.

The routines seem to be well implemented at different hierarchical levels in the school organization. Principal 3 (P3) emphasizes, for example, how all staff at the school have knowledge about systematic quality work. At the school level, the routines have been refined and adjusted to the local needs and conditions. The quality system seems to be quite complex. For example, P3 reflects that ‘there are a lot of systems within the system’, in turn affecting the principals’ work. The tasks required by the system cannot be done by the principal alone but need to be distributed further in the school organization.

The tasks, it’s not only that I can do them myself, but I also must filter them down to my staff in different constellations, in different arenas, capture the information and then pass it on upwards. (P3)

On several occasions each year, the superintendent, and representatives from the LEA hold quality dialogs with the school principals. The dialogs have different focuses, such as ‘the resource dialogue’, ‘the grading dialogue’ (students’ results in terms of grades and merit values) and ‘the school development dialogue’:

Yes, we have for a year the different dialogs in which we meet with the LEA. School improvement dialogue, resource dialogue and grading dialogue, which are good opportunities for each school principal to present different aspects of the school organization, but also to get these questions, as I said before, and that you should be able to explain why. And in those conversations, you identify many things from different schools, and then, it is the task of the LEA to see a pattern in all schools. And you do that sometimes. (Principal 1 [P1])

There is a strong focus on reporting and following up on different types of data and the LEA publishes reports regularly, the so-called ‘cold facts’, which are statistical summaries of students’ grades, merit values, surveys and so forth. The data in the reports are divided into four themes: economy, personnel, premises, and students. The primary aim of the reports is to provide a shared concept of reality that can serve as a basis for discussion, analysis and decision making. Once a year, the LEA also publishes the ‘goal attainment’ report, which is a statistical report that provides how goal fulfillment has developed at each school.

The LEA often makes summaries of things that they show to everyone. Now, we have seen that … After these dialogs we see this … . They make these overviews to create a common understanding, and I think it is important if you want to get everyone involved in the thinking, and perhaps even more important in our organization because we have such a very flat organization. Each school is like its own world. (P1)

In the interviews, the school principals also reflect on monitoring and sanctions, which are parts of Scott (2008) definition of regulative aspects. ‘There might be one and another who think it is too strong control’, P4 says. ‘This control also results in administrative work’ […] ‘You have to be aware of that as well, but it also gives me … I keep a close eye on every single kid at my school’. (P4)

The monitoring system is not appreciated by everyone, and the principals mention various sanctions if you do not do what is expected. This may be reflected in the annual salary review or that you could be summoned to the superintendent and ‘seized by the ear’. The specific school improvement assignments can be regarded as sanctions as well. If principals do not manage to reach good enough results, the LEA intervenes with specific school improvement activities, bypassing the principals and their schools. One of the principals reflects on the fact that the superintendent gives the principals great personal responsibility, which is reflected in the following quotation from the superintendent:

And they know that if they have their student results in order, they have satisfied employees, they have satisfied parents, satisfied students, they have the money in place, then they can do almost anything. Almost. (S)

In the interviews, it appears that all principals in the organization are probably not as comfortable with this responsibility. However, this degree of personal responsibility also creates a great deal of agency:

As long as I produce a result, frankly, I can do exactly what I want, to be honest, for that delegation the superintendent has given me. (P4)

To summarize, the routines of the boundary practice of the systematic quality assessment system are characterized by clearly explicit and well-known regulative elements. Through these routines, a strong coupling between the hierarchical levels in the governing chain is expected to be achieved. By adjusting the LEA’s quality assessment system to their local schools’ contexts, the principals filter it down to the teacher level. A range of quality activities such as quality dialogs, function as routines that sustain and strengthen the connection between the superintendent’s and principals’ communities of practice. There are also boundary objects, such as the calendar, school improvement assignments, statistic data, manuals, and documents such as ‘cold facts’ that function as connections within the boundary practice of the LEA’s quality assessment system. The strong focus on follow-up and monitoring different types of data is not appreciated by everyone, and there are various types of sanctions if the principals do not do what is expected. On the other hand, the superintendent gives the principals great autonomy, though not everyone is equally comfortable with this.

Normative expectations

The boundary practice of the systematic quality assessment system is characterized by the normative expectations of following the rules and doing what is expected, for example, reporting data on time:

The common clear routines and guidelines we have; they are there to be followed, and it is expected of everyone to do. (P1)

One of the principals reflect on the importance of attending the quality dialogs:

P3: I just get an invitation, and then, you go there as well.I: Mm, and then it’s important to be there?P3: Yes, it is important to be there, and it is important to be well read and have done what you are supposed to.

Further normative aspects that are prioritized and rewarded are improved student outcomes in terms of merit values. The superintendent describes how improved student results are what it’s all about:

You build the organization to create pedagogical development, or actually improvement of the students’ goal fulfillment, because that’s what it’s about. [… .] In my case, pedagogical leadership is about creating an organization that leads to goal fulfillment for the students. [… .] All the principals know that if you work for me, then there is only one thing that counts. It is that the students get better. (S)

These normative expectations are well known by the principals:

I: Are you affected by the LEA?

P1: I absolutely believe that.

I: How?

P1: If the LEA says that it is important to have a good goal fulfillment, for example, and sends down demands to me to deliver this merit value or whatever it may be. It is clear that I am affected and that this will be my assignment.

There are also normative expectations for the principals to be strong and energetic as school leaders, along with implicit expectations that principals who do not manage to follow the rules are expected to resign. The superintendent reflects that not all of the principals meet expectations:

I: What is it that makes some people do it and some people do not?

S: It is probably to what extent you have the strength to push for change at the school or, sometimes, at the individual level. Sometimes, you must be satisfied with good enough.

I: So, it really depends on how much the principal manages to push or wants to push it?

S: Yes, and how much I have the strength and will to push toward certain principals. Because that’s where it is, I want something done. So, either the principal likes it and does it, or the principal realizes that I should like that; otherwise, I may not have anything to do in the organization, and then, you do it. And then you also know that we follow it up, that it gets done.

To conclude, the normative aspects, which are prioritized and rewarded, are characterized by following the rules, doing what is expected, improving student outcomes in terms of merit values, being independent and being accountable. The interviews indicate strong normative leadership from the superintendent.

Shared conceptions

In the boundary activities of the systematic quality work, there is a strong focus on students’ academic outcomes, here in terms of merit value:

What I focus on all the time, that’s really the only thing I care about. If you ask me, it’s how high a merit value a student can get. And by constantly turning back to each question: In what way does this activity improve the student’s academic outcomes? [… .] This perception that increased merit value is important and that increased merit value means that students learn more – that is important. We must conquer it every day. So, it’s a language to speak, that we do our follow-ups, that I’m clear about it. We must keep going because it disappears immediately if you do not talk about it and chase it. (S)

This is reflected in the principals’ priorities as school leaders. The shared conception of the importance of measurement is foundational for the organizational infrastructure. The superintendent emphasizes measurement as an important strategy, but not all principals agree:

I: That it is the grade that shows the student’s success? Not everyone agrees with that?

S: Not everyone, far from everyone. But I would like to say that we have managed to change that view quite a bit. So that we have quite a few who … Even if you do not embrace the whole idea with joy, it has been accepted somewhere that this is the way it is. [… .] I like when you measure things, and I believe that a good school is a school where the students perform well. I believe in that. I always get suspicious of teachers or principals who talk about how good everything is, but when I check, I see that it is not so good when it comes to students’ goal fulfillment. I do not think you can be a good school and, at the same time, have bad merit values. I do not believe that.

The emphasis on measurement follows the cognitive element of the importance of follow-up because the superintendent says: ‘I think it’s about follow-up. I think that’s the basis’. The principals also express ‘structure’ as an important cognitive aspect, which is also made explicit in the local policy documents and routines that regulate the boundary practice of systematic quality work:

There is also a lot of structure. I like it. It’s part of my leadership to keep things in order. I like that. (P4)

To conclude, students’ academic outcomes, in terms of merit value, measurement, follow-up and structure, signify the cognitive elements that provide the infrastructure of the school organization.

The context of the regulative, normative, and cognitive aspects, which has been made visible in the interviews, will certainly define the school principals’ daily work. In our analyses, we also found various examples of adaptive strategies of how the principals were handling these expectations. In the following section, the principals’ adaptive strategies are presented under the headings of ‘bridging’ and ‘buffering’.

Principals’ adaptive strategies

Bridging

In the interviews, the principals reflect on the importance of being part of and being loyal to the hierarchical governing chain:

I: So, you are clear that you are staying at your level?

P1: Yes, we are so obedient.

The principals also emphasize how the routines of the systematic quality work has to be integrated in their daily leadership work.

So, our systematic quality work permeates all our work, and for me, it is very important to make it visible in a transparent leadership between the various levels because it is also the case that the teachers are involved in our assignment, and it is they who write the activities we work with to achieve the goals. So, they are completely on track in our systematic quality work, I would like to say. [… .] When you have nagged through it a couple of times, they are on the track now; they actually work with our systematic quality work all the time. (P4)

As part of the governing chain – and to strengthen the connections between the superintendent’s and the principals’ communities of practice – the principals reflect on the importance of reporting results to the superior levels. For the school principals, it is important to have good relations with the LEA and the superintendent because they sometimes need support in their daily work:

The LEA is important to me in my daily work. There are people who are responsible for different areas that I feel I can get help from or discuss with or follow up things with or whatever it is. (P1)

Buffering

Even if the school principals emphasize the importance of using cooperative strategies to sustain and strengthen the boundary practice of the quality assessment system, they also give examples of buffering strategies to protect their schools:

I usually talk about being a filter in both directions, because I am. Sometimes, you have the strength to resist, and sometimes, you let it through, and it’s both upwards and downwards, you understand? (P4)

When we get these school improvement assignments, I do not put everything down at the teacher level. [… .] Some activities, they must be included at the teacher level, but other activities, I can have myself with my assistant principals or just with my head teachers and so on. (P3)

After all, it is important for principals not to undermine the decisions maker at the higher levels. Referring to situations when the principals express some doubt when presenting a decision, the teachers can understand that the principal does not fully agree with that decision.

They understand that I may not really be 100% with this decision. At the same time, I think it’s important that I do not in turn undermine. I think I’m wrong if I say that yes, you know these, the LEA, they come up with a lot of nonsense, you know. The politicians, they do not understand anything, so we are here against the world. I do not think I contribute to teachers feeling involved in this chain in any way. Then, I cannot implement too crazy decisions, of course. (P1)

However, there are still principals in the organization that oppose decisions from the LEA:

I: Do you feel that you are different in the municipality, that some do in one way and others do in another way, so there are both directions?

P1: Yes, it does. Someone is very loyal and always does as you are told and someone thinks no.

I: I do not care?

P1: I ignore that. I hear what you’re saying, but I ignore it. Maybe it’s classic.

The superintendent is aware of this:

That’s it, we always have people flying under the radar, trying to fly under the radar. And somewhere … I have more than 30 principals and schools. It goes without saying that I have three or four that do. (S)

Buffering strategies are used to, for example, make it possible to prioritize working with the students’ well-being and social aspects of education.

We measure a lot in our municipality, for better or worse. We keep a close eye on our students. What we need to work on is norms and values, well-being, and security. I think that can definitely make you go further. (P4)

Students’ well-being, it cannot be measured, so it is difficult to perhaps touch it. The LEA looks at surveys. There is a high level of security, yes, the parents think that it is, yes, it says a certain thing, but then each student’s well-being and some of the students’ personal journey. [… .] You never see that these students who have made a great journey will not be included in this, so to speak. (P1)

I would like to work even harder with a school that is based on getting tools and equipping for life in a multicultural and very exciting world. [… .] But I do not have time to do that because I get to do much of the other things that I am expected to do. I must take that match with myself. But what I think I really do is that I hold on, I hold on to my student health, I work with those parts, and I stand for that, but I could do much, much more. [… .] What affects me negatively is when you do not put the student in the center. [… .] I prioritize well-being above all, and this applies to both students and staff. I prioritize a school that generates goal fulfillment. But if you read on the school’s website, you can see how I write about goal fulfillment, and it’s not just about chasing merit values. It is also to be prepared for life. And I’m the only one who writes about it in the municipality, and I’ve had a lot of fights about this. Do not chase me for merit values. I take care of it, but the other part we also must talk about, but we do not. So, the students’ well-being. (P2)

The superintendent knows well these buffering strategies and finds it quite problematic:

I: The governing chain – what problems do you see making it work?

S: It’s a classic, it’s the whispering game; that is, decisions are distorted. The more levels involved, the more distorted it is. You have that in all organizations.

I: Is there a critical level in the governing chain?

S: The critical level, I think, is the principal – teacher. I can govern, I can control, and I can make great contributions as superintendent. But to know what is said in the next line, it is quite difficult. If I say, for example, now we do this, and it often works that way. We often have an idea, then it is processed in the management team and then it becomes a decision. Then I know that about 20% think that is great, 60% think yes, it’s perfectly okay. And then I have 20% who do not think it’s good. So, it is in all matters. But what I must ensure is that those who do not think it is good go out and tell their staff that now the administration has decided that we should do this. Because then I know that the whole process goes down the drain. [… .] I usually say I want independent principals who, I should not say do as they want, it is completely wrong, but who make independent decisions. They have authority, they have legitimacy, and they must be able to do everything. But there is free play within a defined area, an expression I use to describe how I want an organization to function. And some things you do not step over.

The bridging and buffering strategies used by the principals, and acknowledged by the superintendent, are framed within an agreed context. The principals emphasize the importance of being part of the hierarchical governing chain, among other things, by integrating the quality assessment system in their school leadership activities at their local schools and meeting the expectations of reporting school results and statistics to the higher levels. The principals also give examples of buffering strategies to protect their teachers and students, but also to safeguard the students’ well-being and the social aspects of education.

Discussion

By analyzing the quality assessment system as a boundary practice, we have aimed to identify what regulations, normative expectations and shared conceptions could be identified between the superintendent’s and principals’ communities of practice, along with how the principals relate to these institutional elements. Different objects and activities within the systematic quality work connect the superintendent’s and principals’ communities of practice. The boundary objects found in the boundary practice of the quality assessment system are characterized by clearly and directly demonstrative manuals and instructions. Tight coupling is achieved by the regulations of the quality assessment system. The activities undertaken by these regulations function as routines that sustain the boundary practice of systematic quality work. Although the supporting structures, routines and systematic way of working are important conditions for school improvement, there can be problems if these conditions create limits or even become counter-productive in relation to school principals’ visions of what is educationally desirable (Ståhlkrantz, 2022).

As a critical strategy for school improvement, the use of data has become a key mechanism for tightening the educational system according to a managerial, top-down approach (Coburn & Turner, 2012; Mausethagen et al., 2019; Shen et al., 2017). By using data for governance purposes, national authorities coordinate activities across administrative levels to improve educational quality. As significant drivers and motivators for the use of data in schools, local authorities and school principals are expected to define, decide on and initiate concrete school improvement activities (Mausethagen et al., 2019; Prøitz et al., 2019; Sun et al., 2016). The boundary activities (e.g. quality dialogs) focus on students’ academic outcomes, here in terms of merit value, depending on individual performance. The types of meetings within the boundary practice of the quality assessment system that the superintendent chooses to hold and how these meetings are structured and implemented, will reflect the governing style (Prøitz et al., 2019). Dialogue meetings can be used as school improvement activities to promote productive relations, democratic processes and bridging gaps between superintendents and school principals (Henriksen, 2018). As part of the systematic quality work in Sweden, quality dialogs are common activities that aim to enable participation and dialogue about goal attainment and about the reasons for possible shortcomings (see, e.g. Adolfsson & Alvunger, 2020). In these dialogs, school actors on different levels in the school system meet for a dialogue about school results and school improvement activities; for example, those meeting could be the superintendent, LEA administrators and the school principals. The purpose is to provide the participants with important knowledge about the results at the various schools, but also about what challenges the principals and teachers experience at their local schools. The kind of quality dialogs held by the superintendent in this municipality will likely contribute to shared learning, which can have positive effects on the principals as professionals and improve the schools. The dialogs may then be strengthening and supportive. In addition, there are also regulative and normative expectations within these dialogs that are stressful. It certainly feels uncomfortable to be compared with others within the framework of a results dialogue while not having delivered what was expected. Not being successful as a principal, in respect to improved merit values, will, for instance, have consequences, such as LEA interventions (e.g. school improvement assignments). This might encroach on the autonomy and legitimacy of the principal and can be interpreted as the LEA bypasses the school principals.

When it comes to coupling mechanisms, organizational routines have both ostensive and performative aspects (Feldman & Pentland, 2003; Spillane et al., 2011). The superintendent uses regulations, normative expectations, and shared conceptions to exert influence over school principals’ behaviors and actions and to strengthen the couplings between the LEA and local schools. These elements, although functioning as guidelines for proper behavior, both constrain and enable principals as professional educational leaders. The boundary practice of the systematic quality assessment system is positioned within a context of high priority normative expectations of raising student merit values (grades) that explicitly and implicitly define how school principals should work and take appropriate actions. However, the principals also express agency and opportunities to be able to influence the superintendent, which has been supported by previous research (see, e.g. Addi-Raccah, 2015; Adolfsson & Alvunger, 2020). The school principals in our study acted as gatekeepers, managing various expectations placed on their schools. Referring to Saphier and King (1985), Hallinger (2011) highlights the importance of the principal as a ‘value leader’ and protecting what’s important. Principals’ own personal values will ‘shape the thinking and actions of leaders and represent a potentially useful tool for working with and strengthening the school’s learning culture’ (Hallinger, 2011, p. 129). Acting as gatekeepers and using buffering and resistance as adaptive strategies is not always comfortable, and it can be challenging to balance the fine line between quality assurance, accountability, and control, on the one hand, and professional autonomy, students’ well-being, and a broad understanding of learning, on the other hand. The school principals testify to various sanctions, such as low salary growth, reprimands, or LEA interventions, if not delivered as expected. This can be identified as a culture of a lack of trust, which can be costly from several aspects (Tschannen-Moran & Hoy, 2000). Trust has been proven to be a key resource for sustainable school improvement (Bryk & Schneider, 2003; Robinson et al., 2009). There is a strong statistical link between trust and student improvement, both regarding academic and social outcomes (Robinson et al., 2009).

The strategies used by the superintendent can be considered expressions of managerial accountability (Møller, 2009), where school principals are responsible toward superiors when it comes to delivering measurable outputs. According to the framework of managerial accountability, there is a strong focus on monitoring the improvement of student learning outcomes and holding school principals accountable. Møller (2009) argues that even if Scandinavian school leaders struggle with ‘the tensions of managerial demands from the outside and their own standards for acting as professional educational leaders’ (p. 42), they do not run any risk if they do not pay much attention to managerial accountability. The status of managerial accountability is more like ‘a status of “anticipated future’” (Møller, 2009, p. 42). This is in line with our own previous research (Rapp et al., 2020) and findings that there is a low risk for superintendents to be criticized or lose their position because of poor student results. Unclear leadership and exceeding allocated budgets seem to be riskier.

It can be argued that educational leaders who emphasize managerial accountability will contribute to narrowing the scope of educational goals (Møller, 2009; Ståhlkrantz, 2022) and do not seem to be effective in the long term (Fullan, 2014; Prøitz et al., 2019; Shen et al., 2017). There is a great risk that the focus is unilaterally directed at raising grades instead of serious concern about how to improve education with the broader question of the purpose of schooling in mind. Managerial accountability risks ignoring the critical purposes of education, such as preparation for participation in a democratic society. Aspects such as democracy, social justice and creativity cannot be easily measured and ‘cannot be captured for measurement within this perspective of performance contracting’ (Møller, 2009, p. 41).

Enhanced leadership, focused effort and shared values are common strategies to compensate for loose coupling (Orton & Weick, 1990). The superintendent in the current study emphasizes strong and heroic leadership. In the interviews, the principals express loneliness, each school being an isolated entity and a lack of trust and cooperation in the school leadership team. This aligns poorly with international research stating that school leadership has a greater influence on schools and students when it is widely distributed, with trust as an important prerequisite for school improvement (see, e.g. Bryk & Schneider, 2003; Leithwood et al., 2019; Spillane, 2005). However, it must be underscored that distributed leadership should not be seen as ‘a blueprint for effective leadership nor a prescription for how school leadership should be practiced’ (Spillane, 2005, p. 149).

Henriksen (2018) emphasizes the importance of school leaders as role models and safeguards of ‘the keys to successful dialogue meetings: trust, openness, broad participation, systematic follow-ups, critical questioning, and democratic processes’ (p. 131). Organizational structures, routines and tools are the means through which school leaders act (Spillane, 2005). As a representative of LEA, the superintendent has the agency and discretion to use the systematic quality assurance system to govern schools by a combination of control and professional autonomy. In this context, we return to Orton and Weick (1990) and their argument for the importance of emphasizing the school organization as a loosely coupled system, one with both distinctiveness and responsiveness. Such a dialectical approach to loose coupling allows organizational members (e.g. superintendents and school principals) to engage in practices where they simultaneously act ‘to increase autonomy or responsiveness strategically, in ways that help them to achieve goals’ (Lingard et al., 2014, p. 15). We pose the question that maybe, it is time to give up the dream of the ultimate tightly coupled school organization, like a rational delivery chain, for the more dialectical interpretation of school organizations as loosely coupled systems. If the ambition of the quality assessment system is to promote good education for all students, there must be a broadening in the discussions of what is meant by good education and school improvement and how to combine various aspects of accountability and professional autonomy.

Conclusion

In this article, we have examined the relationship between a supervisor and a sample of four school principals in the context of school improvement. Drawing on an institutional perspective, the aim of the current study was to examine how the superintendent manages the principals through the LEA’s quality assessment system and what consequences are imposed for not reaching standards. Two questions guided the study: i) What regulations, normative expectations and shared conceptions can be identified between the superintendent and principals in the boundary practice of the quality assessment system? and ii) How do the principals relate to these regulations, normative expectations, and shared conceptions? The findings indicate clear and well-implemented regulative routines, but also monitoring and a range of sanctions where the principals do not meet expectations. We found examples of normative elements emphasized by the superintendent, as well as cognitive elements, which provide the infrastructure of the organization. To handle these institutional aspects, school principals use adaptive strategies, such as bridging and buffering. On one hand, the principals emphasize the importance of being part of the hierarchical governing chain and integrating the quality assessment system routines in their daily leadership activities. On the other hand, they function as gatekeepers to protect their schools, students’ well-being, and the social aspects of education. The findings of the study capture the underlying ambivalence in that principals are expected to act autonomously to deliver grade improvement but at the same time tied into a punitive system of control. The study has provided in-depth insights into how global and national policies for quality work in education have been translated into a specific local school context. The small sample is a relevant limitation of the study and makes it difficult to generalize the results. We still believe that the results of the study enable a broader understanding of quality work in different educational contexts.

To be sustainable, school improvement requires consensus and a shared mind-set and that all actors in the school system, at all levels, act on this basis. However, in this collective, mandated endeavor, we need to pose an important question: What is the consensus and the shared mind-set about? In recent years, there has been criticism of the managerial accountability movement, as an expression of new public management (NPM) (Hood & Dixon, 2015). Instead, a turn to trust-based governance reforms can be identified (Vallentin & Thygesen, 2017), emphasizing increased trust in the governance chain, strengthened focus on core business, shared responsibility and trust in professional autonomy (SOU 2019:43). Since trust has been identified as a core resource for school improvement, further research is needed on how the possible policy turn toward trust-based governance will affect quality management in different educational settings.

Disclosure statement

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

Additional information

Notes on contributors

Katarina Ståhlkrantz

Katarina Ståhlkrantz is Senior Lecturer of Education at Linnaeus University. Her main research interests include school governance and educational leadership at different levels in the school system.

Stephan Rapp

Stephan Rapp is Professor of Education at Linnaeus University. His research interest is focused on school legal issues and pedagogical leadership.