Principal’s autonomy and control in Sweden before and during COVID 19 - an explorative study

ABSTRACT

Despite a growing body of research, there is an urgent need for studies on principals’ work during the COVID-19 pandemic. Given that Sweden, unlike most other countries, decided to keep schools open during the pandemic there is plenty to learn from the Swedish case. This article explores how Swedish principals experienced their autonomy before the pandemic and whether and how the experience of autonomy changed over the course of the pandemic. Data from 14 urban principals were integrated in order to analyse principals’ autonomy within four domains of their work. The results indicate that principals experienced a high degree of autonomy in their work, both before and during the pandemic. At the same time, they experienced a significant degree of control from state and local municipal level. The article adds important pieces to the current body of research by showing how autonomy and control impinge upon principals’ work and leadership in times of crisis.

Keywords:

• Autonomy
• control
• Covid 19
• principals
• Sweden

Introduction

This paper focuses on the work of principals before and during the COVID-19 pandemic in Sweden in 2020 and 2021. In particular, we will examine principals’ perceptions of autonomy and control during the crisis. A focus on Sweden is certainly of interest internationally, since, due to constitutional tradition and legal restrictions, Sweden’s COVID-19 ‘lockdowns’ were rather limited, and instead the government allowed for personal responsibility and self-determined behaviour, shaped by public recommendations.

The paper contributes to the growing body of research on school leadership in relation to the COVID-19 pandemic. The stories told share many similarities. There were no guidelines nor prior experiences for school principals concerning how to lead their schools under such tremendously difficult, complex circumstances. In this regard, Harris (2020) claims that school leaders’ work before the pandemic was structured by reasonably stable and common frameworks, which seemed more or less to have disappeared during the pandemic. For example, formal staff meetings and informal conversations in the corridor no longer took place, and principals’ leadership became almost two-dimensional, executed from the computer and/or over the telephone. Still, complex decisions had to be made quickly, with little way of predicting or projecting what the future would look like. In the same terms, Netolicky (2020) describes how the constant changes led to constantly shifting conditions within the school and for school leaders.

Looking further into the current body of research, the work of Hooge and Pont (2020) addresses the importance of school leader responsiveness and flexibility in unpredictable times, which indeed became evident during the COVID-19 pandemic. For instance, this implies that courage and commitment are needed when facing complex situations, dilemmas and conflicting interests. The work of Jarvis and Mishra (2020), focusing on how thirteen leaders at schools, colleges and universities dealt with the pandemic situation, also revealed that school leaders were undoubtedly leading during the pandemic, even if they did not exhibit all of the classic signs of crisis leadership. For example, they remained instructional leaders but in circumstances radically different from those imagined by any literature on the subject. In the more recent work of Oplatka and Crawford (2022), it is emphasized that in the reopening of schools and the return of school staff and students to their ‘normal’ educational settings, school leaders should pay great attention to the emotions that the pandemic (and post-pandemic) crises have caused, for instance by supporting their staff in coping with a sense of loneliness and frustration many of them experienced during this period.

The body of evidence is growing, but there is still too little known about how school leadership was executed and navigated during this extremely challenging and uncertain period. Focusing on this will help us develop profound knowledge for handling future crises. In that sense, Sweden provides a noteworthy and important example, an important case to study since, in contrast to most other countries around the world, Sweden chose a different path to deal with the crisis. More specifically, Sweden chose to keep schools open throughout the pandemic, and instead of a lockdown, principals, teachers and other staff were required to closely follow the directives and recommendations that state and municipal authorities provided. It is also worth emphasizing that, in the later ‘pandemic waves’, some countries chose to pursue a path closer to the Swedish one.

Pioneering research has started to broaden the understanding of principals’ work during the pandemic. The work of Ahlström et al. (2020), for example, identifies three areas that were particularly challenging for Swedish principals under the strategy chosen by the Swedish government during the pandemic. The first area identified was the concerns of students, parents and school staff. There were clear instructions to all these groups to stay home in case of symptoms, which led to increased absenteeism at schools. For those teachers and staff who remained on site, the workload increased. The second area identified was the constant state of uncertainty that prevailed. The variability of the virus gave rise to routines often being reformulated at the same time as guardians demanded more information – information that the principals did not have when they had no more information than was available to the rest of society. The third and last area concerned students who ‘fell behind’ in their school work. Many students need a clear structure to function properly in their everyday life and to be able to handle the school’s requirements. Obviously, many such structures disappeared with the COVID-19 pandemic. Ahlström et al. (2020) revealed that many principals felt great concern for these students during the pandemic. With regard to the Swedish case, important questions emerged around school system governance but also around (experienced) autonomy at different levels of the system, given the exceptional situation that evolves during a pandemic.

Given the factors presented above, previous work indicates that principals in general, and Swedish principals in particular, had to navigate many complex situations in their leadership during the pandemic. Arguably, and linked to the work of Cribb and Gewirtz (2007), important questions concerning autonomy and control were actualized because autonomy might not always be desirable, and control or state governance can have a negative influence. In order to attain a deepened understanding of these questions, the article takes its theoretical point of departure in previous work of Wermke and Salokangas (2021), and Wermke et al. (2022) in which teachers’ and principals’ experienced autonomy is divided into four domains (detailed below).

Against this backdrop, the aim of this article is to explore how fourteen Swedish principals experienced their autonomy before the COVID-19 pandemic, and whether and how the experience of autonomy changed during the pandemic. The following research questions directed the empirical work. A two-fold perspective on autonomy before and during the pandemic is important, in order to render visible changes in school leadership resulting from crisis.

1. How do school principals in Sweden describe their autonomy and control?

2. How have school principals in Sweden experienced that autonomy and control to have changed throughout the COVID-19 pandemic?

The article at hand investigates principals at both secondary and upper secondary schools. Upper secondary schools shifted their instruction to remote forms, while all others schools, such as compulsory schools, remained open. Therefore, it is important to state that whether schools were kept open or closed is not central to our study of autonomy. Rather, the key issue is principalship in times of crisis, with a particular focus on the differences in decision making and control, compared with what has been seen as normal for principals in Sweden, especially in terms of the complex relations of principals to their teachers, educational authorities, students, their parents and the surrounding civil society. However, in a few cases, relevant differences between upper secondary school and elementary school principals appear and in these cases the potential differences between school forms are also noticed.

The article is structured as follows: First, a brief overview of the Swedish school system is presented. Then, the theoretical point of departure is described. Thereafter, the analytical work is detailed, followed by the results. The article ends with a discussion, conclusions and some suggestions for future research.

The Swedish education system and Covid 19

In the post-war period, Sweden has been described as one of the most centralized school systems in the western world (e.g. Jarl et al., 2018). In this system, the government provided detailed directives to the local political level, i.e. to the municipalities, for example, regarding employment of principals and teachers, salary standards, distribution of financial resources, the content of the curricula, and so forth.

The 1990s, however, became an important watershed moment in school governance and principals’ work in Sweden. In that decade, decentralization and marketization reforms were implemented in Sweden (Lundahl, 2002a;2002b), and in this shift, management by objectives and results was replaced. Detailed state regulation and the responsibility for organizing public education, and for allocating resources to schools, was transferred from the state level to the 290 municipalities. The National Board of Education was replaced by a new government agency, the Swedish National Agency for Education,¹ which adopted a principle of ‘stopping at the municipal border’, to uphold the freedom of the local education authorities (LEAs) (Jacobsson & Sahlin-Andersson, 1995). In short, during this decade the Swedish education system was transformed from one of the most centralized into one of the most decentralized in the world.

Moreover, the school system was turned into a school market, by opening up parental choice through school vouchers, and for non-profit and for-profit private school owners. This considerable local freedom, however, has led to significant problems in the Swedish school system, in turn leading to results crises in relation to decreasing PISA results, and to legitimacy crises in relation to decreasing social justice. Consequently, there has been a change in Swedish school governance in terms of a re-centralization movement, where the state is more present at the municipal level compared to the 1990s and early 2000s. The establishment of a Schools Inspectorate in 2008 marked an important step in this development (Rönnberg, 2012). A primary task of the School Inspectorate has been (and still is) to inspect principals’ work in relation to national regulations, for example, regarding the schools’ systematic quality work, equality, grading and assessment. In addition to external inspections and an increased number of state directives, a number of large-scale school improvement initiatives were launched by the National Agency for Education in later years.

Consequently, the responsibility for education in Sweden is today divided, roughly speaking, between three political levels and several actors: the parliament, the government and the government agencies at the national level, local educational authorities (LEAs) and independent school providers at the municipal level, and schools, principals and teachers at the local school level (Wermke & Salokangas, 2021). In the 290 municipalities, public education is directed by a school board and a superintendent. Rather characteristic of Sweden, the school boards consist of appointed politicians representing the political parties of the municipal council. The whole system is today closely supervised and audited by national authorities. It can be described in terms of extended local decision-making (autonomy) but also by extended state control (Wermke et al., 2022). This configuration encountered the crises caused by Covid 19, hitting the world at the beginning of 2020.

As noted earlier, Sweden chose a different path in the COVID-19 pandemic to many other countries. In Spring of 2020, the government introduced a temporary regulation giving principals at secondary and upper secondary schools the opportunity to decide on, for example, remote and distance education and changed lesson times (for holidays and weekends) (Swedish National Agency for Education, 2021). However, one practice was established very quickly. Upper secondary schools shifted to distance education, while other Swedish schools remained open. These schools had to operate under specific legal frameworks not found in ordinary school legislation. Recommendations were passed down from the Public Health Agency of Sweden stating that students and staff should stay at home if they had symptoms of Covid-19, and that infection tracking should take place in the event of confirmed infection at a school (The Public Health Agency of Sweden, 2021). Principals were given instructions, for example, concerning risk assessments for students and staff in risk groups, as well as a special risk assessment for pregnant women. At the same time, it was the school boards’ (either the LEAs or private school owners’) responsibility to ensure that measures were taken to reduce the spread of infection at the school and if a student or member of staff was exposed to COVID-19 infection at the school, it would be reported to the Work Environment Agency as a serious incident.

Against this backdrop, principals’ autonomy and the control of their leadership actions becomes relevant to explore, especially in a country that was not ‘locked down’, in relation to new principal work tasks that had to be handled, often at the cost of other, usual work tasks, and in relation to decisions that had to made without a mandate, phenomena that are known as ‘autonomy gaps’.² Moreover, even if the government, its agencies and the municipalities or their independent counterparts send directives to principals on how to navigate the pandemic, the decentralized model for governance still demanded that principals, as ‘street level bureaucrats’ (see e.g. Jervik Steen, 2014; Lipsky, 2010), had the responsibility for many decisions at the local school level. In that sense, it is relevant to explore how they experienced the autonomy and the control exercised, both from the state and the school owner level.

Theoretical perspectives

The article employs Ingersoll and Collins’s (1996; Ingersoll, 2003) and Ingersoll and Collins’s, (2017) theories of power distribution and control. Their work also covers the work of principals. Decision-making abilities are seen in relation to the degree of conflict in the school, and autonomy is here seen as the possession of a high degree of control over issues concerning daily activities. Ingersoll (1996) states that a school actor’s capacity to exercise autonomy has a positive effect on school function, but that the effect is dependent on which activities can be influenced.

Regarding Ingersoll’s (1996; Ingersoll, 2003) and Ingersoll and Collins’s, (2017) ideas of power distribution and control, decision-making abilities are seen in relation to the degree of conflict in the school world, and autonomy is here seen as the possession of a high degree of power over issues concerning daily activities. Ingersoll’s approach is beneficial for cross-cultural studies, due to the facility of connecting his theoretical considerations to a very straight-forward and context independent operationalization. Ingersoll formulates four empirical questions that guided the study design: (1) What are the most important decisions to be made in relation to principals’ work? (2) Who is allowed to make such decisions? (3) On which foundations are these decisions made? (4) Who ensures that the decisions made are appropriate? Questions 1 and 2 relate to the autonomy of principals concerning the most important decisions and whether the principals are allowed to take these decisions. Questions 3 and 4 relate to the control dimension, which for us means detailing how the most important decisions are constructed and evaluated.³

Based on this theoretical framework, the analytical work is structured in terms of an analytical device for the empirical investigation of autonomy in education from a comparative perspective. This device, which has been applied and verified in previous works (e.g. Nordholm et al., 2022; Wermke & Salokangas, 2021; Wermke et al., 2023), builds on the idea of autonomy in education as a multidimensional phenomenon. This means that actors in school systems can obtain and/or lack different forms of autonomy in different dimensions of their profession. Building on this device, the article will investigate principal autonomy in terms of the different ‘domains’ in which decisions can and must be formed. In relation to the work of Wilches (2007), of Ingersoll (2003), Lortie (2009), but also the classic study of Rosenholtz (1989), concerning the workplaces of principals, the article proposes four different domains which are essential to principals’ work. The definition of the domains presented below may be open to discussion, but above all our grid attempts to capture the multidimensionality of autonomy in public education.

(a) Firstly, by educational domain the analysis refers to matters related to activities and responsibilities connected to teaching and learning. (b) Education and schools more specifically play a crucial role in the socialization of students. This domain is termed the social domain. Examples of such processes would include, for example, grouping students, either randomly, or based on their gender, ability or developmental stage. Another example of socialization would be fostering students according to certain norms. (c) Developmental domain refers to decisions that relate to identifying and steering the school towards a ‘vision’ or a plan of action. This domain refers to what principals plan and do in order to develop the school, and in steering the direction of the school in matters such as professional development of staff decisions. (d) Finally, by administrative domain, the analysis refers to the administrative work of schools that facilitates learning and other possible activities in schools, distinguishing educational and administrative duties of teachers, including decision-making concerning, for example, timetabling, use of resources etc.

Following this device, in relation to the focus on how Swedish principals experienced their autonomy before the COVID-19 pandemic and whether and how the experience of autonomy changed during the COVID-19 pandemic, the article categorizes 1) what principals see as their autonomy and responsibility in the most important decisions in their daily work and 2) to whom they feel accountable in their work regarding the most important decisions to be made. The following section details the sample, the current case and the data analysis undertaken.

Sample and data analysis

Empirical data were collected to demonstrate certain aspects of how Swedish schools’ principals experienced their autonomy and control before the COVID-19 pandemic and how they experienced changes during the pandemic. In this article, interviews with a total of 14 principals were conducted (n = 14). The principals worked at both secondary (n = 8) and upper-secondary (n = 6) schools in the middle and northern part of Sweden, however all in urban areas. Eight (n = 8) were men and six (n = 6) were females. For this article relevant, all principals had at least five years of experience but some of them had worked much longer as principals. Moreover, the total sample have important similarities regarding the municipal school organization given the research questions put forward. For instance, all principals had access to central support from the central municipal level in terms of HR, economists and centrally support for student health, and so forth. However, in a few examples, the specific contexts and/or school form rendered in clusters of respondents, which is also noticed in the below-analysis.

The interviews were conducted between Autumn of 2021 and Summer of 2022. Arguably, despite the common features detailed above a sample of 14 principals is somewhat limited, and therefore we want to emphasize that the study at hand is of a qualitative nature, with the ambition of making analytical generalizations on principal autonomy in times of crises, i.e. of making a conceptual contribution that could be validated in further empirical studies at scale.

Regarding coding, the interview transcripts were analysed deductively, based on the four domains presented above. Within the educational domain, interview transcripts targeted issues related to organization of teaching (e.g. planning, instructions, implementation and evaluation). Extracts addressing the school’s governing documents, i.e. the Education Act, regulations, national curricula were also covered in this domain. In the social domain, extracts targeted socialization of students, grouping of students (randomly or by level, gender, and so forth), disciplinary matters, helping students with special needs, student health work, guaranteed equal treatment and student well-being from a principal perspective. This domain also comprised feelings related to the pandemic in students and their caregivers. Extracts addressing guardians’ influence on principals’ decision-making also belong to this domain. Regarding the developmental domain, extracts were linked to the principal’s vision or pedagogical goals, extracts about strategic development, how the school is managed and matters of continuing education. Extracts detailing school development, quality work, development processes, school culture and development plans were also coded into this domain. Finally, the administrative domain addressed extracts on human resources, staff recruitment, budget, scheduling, salaries, composition of the timetable, routines, division of duties for teachers and property issues. Infection tracking and was also coded into this domain.

The results presentation below is structured in terms of the four domains. Each section is also structured in relation to the two research questions put forward addressing how principals perceived their autonomy before the pandemic and how or whether it changed during the pandemic.

Results presentation

Educational domain

Principals’ experienced autonomy before the pandemic

All principals described themselves as being governed by the juridical and pedagogical directives – mostly by the Education act to directives and recommendations from the government’s school agencies. One principal, Erik, explained how laws and regulations can regulate and control his work, but that at the same time, they also define and clarify his assignment and the autonomy offered. David had a similar line of reasoning, saying that laws and regulations certainly govern his work but that he does not automatically perceive them as a form of control of his work. On the contrary, he mentioned that the contact he has with the Swedish National Agency for Education is something positive, i.e. support in his everyday life. Christian, another principal, considered himself to have ‘very far-reaching authorities based on the Education act’ and, on direct questioning, he considered himself to have great autonomy within his role as principal.

Our analysis also revealed examples in which principals not only detected their autonomy within the framework given, but sometimes also explored their autonomy and tested their boundaries. John, who had extensive experience of working as a principal, expressed his autonomy linked to pedagogical work in the following way:

You also, with your experience, give yourself a little more freedom all the time and test the limits. And I have probably said at some point that if this promotes the school’s work and the students’ learning, I’m prepared to at least step over [the limits] and see what it looks like on the other side of the legal text and test it.

Regarding autonomy and control framed at the local municipal level, principals explained that they have a high degree of autonomy within this domain. For instance, Fredrik described how the municipality sets some overall goals for all upper secondary schools, but that these goals are quite broad, giving principals and individual schools a large scope of action. Another principal, working in a socio-economically prosperous area, Jack, declared:

It [the autonomy] is huge. Because we have a school with a high merit value [student results], we can do things more or less as we want. However, then you are sensitive to customer satisfaction and I think you should be … But the freedom to do as you wish is considerable as long as no one complains to the municipality.

In other words, there are evidently benchmarks in relation to the academic results of your students. As long as your school ‘delivers’, your autonomy as the head of the school can be very extended, a factor which is very much, and not surprisingly, related to the above mentioned output governance rationale inbuilt in the Swedish school system. In relation to this notion of governance, and as also demonstrated in other domains below, our principals also stated that they simultaneously experienced an extended degree of control from the central municipal level. Accordingly, and unsurprisingly given the theoretical point of departure, it becomes clear that the concepts of autonomy and control should not be understood as poles of a dichotomy, but rather they can be complementary.

Principals’ experienced autonomy during the pandemic

As mentioned earlier, principals in Sweden had very little autonomy regarding decisions concerning whether or not instruction was provided remotely. Here, the principals were given clear directives, both from state and municipal level, upper secondary schools provided instruction at distance, all other schools had to remain open. Our principals experienced this as a restriction of their autonomy. An upper secondary school principal, Fredrik, explained:

the possibility of making a difference for the student [diminished], introducing part-time distance learning makes it worse for the student, introducing full-time distance learning makes it worse for the student, not being able to carry out any development work makes it worse for the student.

Such an experience of restricted autonomy is interesting, since before the pandemic, principals could not make decisions about having certain students study remotely, while others did not. We argue that rather the entirely new premise of distance schooling was challenging for the upper secondary principals. For our secondary school principals there were few changes or none at all. Petra said:

And the educational aspects were still left open, the ‘how’ of instruction. This was left. Indeed, even more regarding how to teach digitally.

Erik argued that distance teaching created an increased workload, but more for the teachers than for him, and that he therefore did not believe his autonomy was reduced within this domain. Björn, however, explained that his autonomy was reduced due to the fact that completely new tasks arose as a result of new ways of conducting learning activities during distance learning.

Another point concerns David, an upper secondary school principal. He mentioned that when the teachers were requested to work from home, he no longer had any control over their working conditions. In that sense, he felt that the teachers’ autonomy increased at the expense of his own and other principals’ control of the schools’ common pedagogical work.

Even in the relationship between the principal and the state administration, our analyses reveal interesting phenomena in terms of autonomy and control. In Sweden, the School Inspectorate regularly inspects both municipal and independent school operations. There are different types of inspection, for example based on how schools have passed previous inspections. However, some parts of the School Inspectorate’s school monitoring were paused during the COVID-19 pandemic, something that principal Christian mentioned in his interview, which, according to him, led temporarily to a higher degree of autonomy. However, in other examples, control within this domain was described as relatively unchanged. Björn, for example, received inspection from the Schools Inspectorate during the ongoing pandemic. However, the inspection was carried out online and it had no connection to the ongoing pandemic. Thus, Björn felt that the state did not relax control within the educational domain in the way that principal Christian described. The analysis also revealed examples which indicated that the Schools inspectorate had become an institutionalized part of the Swedish school landscape and thus the short break during the pandemic did not affect principals’ autonomy and control to any great degree, as here described by Lena:

And the rigorous control. I participated in two inspections this Spring. One in my old school and another in my new independent school. I must say it was like an interrogation. It was almost a court-like process. So this inspection, it really is control.

In summary, autonomy is generally perceived to be high within this domain, and this autonomy was felt to be reduced during the pandemic, mostly for the upper secondary school principals, due to the increasing challenges associated with distance education and the lack of oversight of teachers. However, it also emerges in the material that principal autonomy remained rather unchanged. The major factor was that the site of schooling had changed for some principals. Also, the control dimension did not actually change significantly. Even if the Swedish School Inspectorate did not visit schools physically, the perception of control was remained for our principals.

Social domain

Principals’ experienced autonomy before the pandemic

All principals stated that they have a great deal of freedom to shape their organization to meet the needs expressed by teachers, students and parents. Several principals emphasized that the most important decisions they make are the decisions that are in the best interest of the students – that is, that all decisions they make should benefit the students and also the teachers. Principal Björn described his role and the purpose of his decisions in this domain:

[My decisions should] empower the teachers so that they can work with the students in the classroom, to remove obstacles to learning in different ways.

Analogously to the previous domain, some principals expressed that, in this domain, there is possibly greater autonomy to act and that one must dare to push the boundaries a little, here exemplified by Fredrik:

Our task should be to exploit and try to expand that autonomy as long as it promotes the well-being and development of our students. As long as it is in the best interests of the students, I dare say that it is our duty to be out there and push at the boundaries.

Another overall impression concerns the degree of control and the source of the control. In contrast to the previous domain, the principals did not claim that they experienced similar degrees of control from the state and municipality, although there is also a clear framework here to relate to, manifested in the Education Act. Instead, the principals stated that a large part of the control within this domain comes from guardians and parents. Lena detailed:

We have all eyes on us in society. The parents control a lot, guardians in this macronized school. They see themselves as customers and the customer is always right.

In similar examples, Björn described how guardians called him directly and made demands and suggestions about how he and the school should think and act. Another principal, Christian, mentioned guardians who called him up and read the Education Act to him in order to have certain demands met. Principals explained that this lack of trust and autonomy was a source of stress and sometimes also irritation. For instance, David said that ‘[guardians] sometimes have valid concerns, but they usually speak without knowing a damn thing’. The significance of the control of parents and civil society, in particular for the social dimension of schooling in Sweden, therefore becomes clear.

Principals’ experienced autonomy during the pandemic

Regarding the pandemic period, principals claimed that they continued to have a high degree of autonomy and experienced a low degree of control, from both the municipal and the state level. In this context, the upper secondary school principals expressed relief that the students from the Autumn semester of 2020, to a certain extent or in full, were allowed to be physically present at the school. This eased the workload with the social tasks of the school’s assignment, particularly concerning the well-being of the students. Physical meetings in schools gave teachers and principals the possibility to, with some restrictions, arrange physical meetings and discuss issues that may be more difficult to talk about virtually. In many schools, for example, the groups responsible for social questions and student health met physically, as Mari acknowledged.⁴

However, efforts of trying to continue to keep students in schools were not always appreciated by guardians and parents. Therefore, the control mostly came from parents and guardians during the pandemic. For example, David stated that guardians questioned regularly about why the students at his school had to be on-site, why the students had to go to workplace-based learning, or why students who travelled long distances with public transportation would have to be on-site at school. He described how a lot of work time was spent explaining and defending the strategy that existed. This illustrates the degree of responsibility principals had to bear during the pandemic. Being on-site in the middle of the storm, rendered them responsible not only for their own health, but also for the health of students and their families. Again, Mari stated:

Indeed, at the beginning [of the pandemic] it actually was about life, severe sickness and death, it was terrifying.

This anxiety was quite real. During the pandemic, media reports of outbreaks in schools, leading to teachers becoming seriously ill or dying, were not uncommon.

Moreover, in our data, we see also the temporal dimension of the pandemic and its impact on principals’ work and autonomy, coming from various governance sources. More specifically, principals claimed that the amount of control from the municipality could be described on a sliding scale from the start of the pandemic in March 2020 until the Autumn semester of 2021. In the initial stage of the pandemic, principals voiced that it was the municipal school administration which straightforwardly decided that all students should be taught through distance learning. However, the longer the pandemic lasted, several principals felt that autonomy increased and that more decisions regarding possible distance education for students were left to the respective principal. Christian explained:

Right here at the end, then it was a lot for the principal to decide, and to some extent I feel that the principal had to find solutions. (to all problems)

Another principal, Mary, expressed a similar opinion, illustrating how autonomy is not always desired by local professionals:

The administration did not say anything, so the school principal made some decisions and said exactly how it would be. And there I can say that it would have been nice not to have to make these decisions.

This finding exemplifies the working conditions of Swedish principals, who must take many decisions, i.e. they have a great deal of autonomy. In other examples, Erik also described how principals navigated areas in which there was not much autonomy on paper. The routine in the municipality, in case of suspected local outbreaks of infection at a school was, for example, that the principal would inform the municipal school administration. They would then contact the infection control in the region to discuss whether, for example, an entire school class would be sent home for distance learning. Erik described, among other things, that he did not always want to wait for such decisions, but that it was better to act and have the discussion afterwards:

When we had a risk of a local outbreak of infection […] I did not wait for information. I informed [the superintendent] and so I acted regardless of whether there was a possible local outbreak. [or not]

In other words, a decision to send an entire class home, for example, was not the principal’s decision to make, but the superintendent’s. However, Erik not only managed to identify a certain room for maneuver, but also managed to avoid the superintendent’s control.

Summing up our findings for principal autonomy in the social domain, the analysis showed that principals experienced a high degree of autonomy and a fairly low degree of control, from the state and municipal level, both before and during the COVID-19 pandemic. The strongest control of principals’ autonomy instead came from guardians and parents, which we identify as examples of relieving the administration of responsibility and making principals accountable to the civil society. The analysis also revealed examples in which principals had to challenge existing directives, for example by deviating from the decision-making paths formally specified by the municipality, which are illustrative examples for autonomy gap management, related to high personal and professional risk.

Developmental domain

Principals’ experienced autonomy before the pandemic

Within this domain, we see a variance of perceived autonomy, which appears to be mostly related to different forms of municipality leadership, which can be trustful of the principals’ abilities as educational leaders, or can be fragmented, disorganized, and top-down. The size of the municipality also plays a significant role. Björn, for example, claimed that school improvement work was the area in which he had the most autonomy in his leadership. Christian and Jack, for example, highlighted an acceptance by the municipal school administration that schools have different needs in their school improvement work. However, Christian also described a specific situation in which a directive from the municipal school administration required all schools to address specific improvement focuses, even though principals already had chosen priority areas. Christian stated that the ability to work on additional issues was impossible for several reasons, something that the municipal school administration quite soon understood:

It was a very clear directive that came at the eleventh hour. It was absolutely impossible for a school to adhere to them. It was a fairly comprehensive task that all schools had to complete. But they [the municipal administration] come in August and give the forms for it, it’s actually completely insane.

Moreover, principals had different opinions about being part of a large municipal administration and how this affects the degree of autonomy and control. A principal, Anders, highlighted the large organization as a limitation of autonomy. He said:

Here we are quite a few municipal upper secondary schools and that is a strength, but it also means you don’t have the scope of action you want. […] You have to consider each other and other things.

Principal Björn sees the matter in the opposite way. More specifically, he believed that a larger organization, instead, creates opportunities, for example because it enables him to reason with many colleagues about problems and to find solutions with them. Another thing, according to him, that creates room for maneuver is the opportunity to receive support and feedback, compared to a smaller school administration, and in this way you can also avoid control. Björn explained:

I like the concept of a larger group. […] In a large organization there are many more opportunities than in a smaller one. […] In a larger municipality - there is much more, where there are many more who have the same role, where you can share more good examples or solutions to more difficult examples.

In terms of control, the principals rarely talked about state control when it came to school improvement work. Regarding the control from the municipal school administration, all principals believed that the ‘unit analysis’ was the most important control function that the municipal school administration exercised. However, all six principals had a positive view of the analysis being carried out and about the dialogue they had with the superintendent. The unit analysis is written by each principal once every academic year. Principal David explained:

My attitude is that it [the unit analysis] benefits my school’s work and us as leaders. […] Put simply, it requires us to reflect when we put words and thoughts on paper.

Thus, developmental autonomy before the pandemic appear to mirror the decentralized nature of the Swedish school system, with various leadership styles in the municipalities and also various sizes. This correlates with the control which principals perceive.

Principals’ experienced autonomy during the pandemic

Almost all of our principals described how their schools paused all work within this domain, here detailed by Mari and Petra:

Mari: Our working tasks were now to organize vaccinations, Covid tests, contact with the public health authorities, and so on. Petra : [The] unions were after us. It became more and more. Mari: Yes, school improvement work did not have the same quality at all.

This does not mean that there was no development work at all. Fredrik described how his school went from working with school improvement to ‘pandemic improvement’, working from day to day with making schooling possible in very uncertain conditions. The reason for this shift was the regulation on distance education that came in the Spring semester of 2020. Principals who were affected by this decision expressed how focus shifted from working with long-term school improvement to solving the teaching situation right here and now. This does not mean that pandemic improvement did not also lead to improving working conditions. All of our principals mentioned the advantage of having meetings online. In particular, bigger meetings, and meetings with people outside the school, such as parents or municipal authorities, have remained in many cases virtual, up until today. Here is Eric’s distinction between meetings and work:

Sometimes, I thought, it was very nice to have a meeting in which we do not have much interaction online. The advantage with having these remotely, is that you actually can work a little bit, when you are listening to information. Fantastic in many situations.

Another important aspect was the professional development of teachers, individually and collectively. Here things happened. In particular, this concerned the challenges of distance education at scale, including necessary ICT profiency. Principals, for example, described that ‘stable’ teachers, with a strong identity, managed the improvement work better during the pandemic, compared to those in need of continuous support. We might argue here that the pandemic, with its challenges, was also a litmus test for teachers’ capacities to adapt their teaching to different forms of change.

Teachers with the capacity to manage themselves have managed this better. They have managed to lead themselves in the right direction… Teachers who, on the other hand, do not have this ability, have ended up behind or isolated… So it has been tough. It’s more unequal now I think.

Overall, principals did not experience any particular control, either from state or municipal level. But here, as well, the analysis showed some interesting nuances. Principal David, for example, described a positive consequence for him that made a difference to his school, giving him more autonomy:

We have gained influence over the requirements and wishes of a small school. […] People [the municipal school administration] have listened to it [our needs] and have not been so rigid. You experienced that in the beginning, that oh my god, everyone has to always do the same.

In relation to our description of some municipalities being rather fragmented and confusing in their leadership, we might argue here that COVID-19 may also have become a useful strategy of exit from inappropriate expectations placed on principals, without loss of face.

To summarize the developmental domain, the analysis showed that, before the pandemic, principals perceived the degree of autonomy to be somewhat high, while at the same time, some principals perceived a high degree of control, specifically from the municipal school administration. During the pandemic, in which almost all development work was paused, control also ceased, something that some principals perceived gave them increased autonomy, despite increased pressure generally.

Administrative domain

Principals’ experienced autonomy before the pandemic

Several principals mentioned that they have a great deal of freedom when it comes to making schedules for each program, the distribution of teachers’ tasks, which courses the teachers will teach in the coming school year, which work team they will belong to, and so forth. However, the critical point of real perceived autonomy, for the principal is state budgetary issues. Both Mary and Fredrik explained that one area where their autonomy had decreased most in recent years was in the financial area. Moreover, as in the developmental domain, we see many differences between municipalities as result of the considerable decentralization. Anders compared his current situation with previous principal jobs in other municipalities and explained:

I have worked in various municipalities and the financial management in one municipality reduced the budget. I have experienced it a lot here in a way that I have not experienced in other municipalities. […] In this municipality, after all, you have 54.7 % services that you can outsource and there is a much more detailed control. […]. The steering is much, much sharper here.

The principals consistently came back to the question of control over budget. One uniting feature in these stories was the desire for less control from the municipal school administration. David described this as follows:

[I have the most autonomy] when I have my money bag here and want to distribute and lead. [The economist in the municipality] ties the hands of us principals. He is not an educator, he is an economist. […] I am being held back because we get a distribution based on which courses we run and not based on what reality looks like.

Another type of control that principals described in more positive terms was the monthly follow-ups that the principals have with their superintendents. The focus of these meeting was, for instance, on administrative issues and routines, as well as how the schools perform. All principals described these follow-ups as something positive and claimed that they help in their work as principals.

Regarding state level control within this domain, the Education Act (2010:800) stipulates that principals are legitimized to lead and distribute the work within their local school organizations. Put differently, principals’ autonomy is formed and empowered from within their own internal organization. However, and as shown above, principals’ autonomy is affected by the control instruments designed by the local school administration.

Principals’ experienced autonomy during the pandemic

Almost all principals experienced a difference in their administrative autonomy during the COVID-19 pandemic. They also described a rather confused state, in which they themselves were expected to solve and administer a long series of questions and problems, as we shown earlier. Christian detailed:

After a while, it was almost as if there apparently was the expectation that the principal and schools would solve things that the rest of the world failed to solve. That is to say, to prevent the spread of infection while keeping us open.

The confusion about what actually applied to the schools and its principals was something that principals reiterate in their accounts, emphasizing that they were left alone and forced to interpret how they should act. Principal Erik voiced this frustration:

During the course of the process, it has been even more difficult this year to transmit messages. Because the information we received (from state and municipal level) has sometimes been very contradictory.

In order to address this situation, the superintendent with responsibility for upper secondary school introduced a weekly meeting to discuss how the week had been, and how to plan for the next. Several of the principals, including David, Anders and Christian who all worked in the same municipalty, identified this as something positive in their context – that they were united and that they made decisions together, which indeed decreased the responsibility the of individual in relation to the circumstances, including all the related complexity and risks. David explained:

There was much closer follow-up on everything. […] Much of the meeting time, at least during half of the pandemic, was spent over hours on ‘what should we do?’, ‘what can we do?’, ‘how should we interpret this directive that has just come out?’

However, a closer look at these meetings indicates that the administration first of all tried to fight the pandemic by employing bureaucratic means. One very significant issue during these weekly meetings was, for a long time, the infection tracking that the principals became responsible for carrying out in their schools. This directive came directly from the Public Health Agency of Sweden. More specifically, this job involved mapping contagion at the school, but all principals called it contagion tracking. Christian explained:

if we had had an outbreak of hepatitis then no one would have come up with the idea that ‘but we should put this on the schools to trace infections’. This suddenly became the school’s task. Fairly complex things to solve I would say.

From a similar viewpoint, Anders commented on his own workload during the COVID-19 pandemic, exemplified by infection tracking:

I did all the infection tracing. […] I spent over 200 hours just with infection tracking and that’s not 200 hours I had left over. No, the workload was gigantic.

In other words, principals in Sweden, had other authorities to serve, in particular the Public Health Agency. This changed their work away from an educational focus. It is important to remember that the work of testing and tracking in the open schools in Sweden had to be delivered on top of the regular workload. The municipalities mostly did not support their principals in this work, in practical terms.

In summary, principals experienced substantial challenges within this fourth domain, which they also had to solve within the scope of the autonomy given. In this respect, regular meetings organized by the municipal school administration and its superintendents were also helpful, according to the principals, but were, however, far from enough. Moreover, state control, mostly exerted by the Public Health Authority, increased, and principals’ autonomy was more restricted. In addition to this state control, principals also had to consider directives from the Swedish Work Environment Authority. Finally, principals were also assigned new and complex assignments, which took a lot of time and thus also reduced their autonomy.

Discussion and conclusions

The aim of this article was to explore how Swedish principals experienced their autonomy before the COVID-19 pandemic and whether and how the experience of autonomy changed during the COVID-19 pandemic. In so doing, we examine the contextualized work of school leadership in times of crisis. This also illuminates which aspects of school leadership are stable and resistant against great challenges, such as a worldwide pandemic. Given that Sweden, in contrast to most other countries, remained largely open, including most of their schools, there is plenty to learn from the Swedish case, especially in terms of research and practice of school leadership.

To start with, Swedish principals, in general, experience a high degree of autonomy in their work. In other words, they can and must take many important decisions in relation to their schools’ organization. At the same time, they experience a significant degree of control from state and municipal levels in the educational, developmental and administrative domains. In the great crisis presented by Covid 19, we hoped we might be able to identify the characteristics of school leadership which remain untouched, become stronger or disappear in a case such as Sweden. To start with, it becomes relevant to pay attention to the level of control exerted during the pandemic and how this indeed impacted principals’ work. It is important, for instance, to move on from normative ideas that autonomy always is considered to be preferable for and from professionals and control is not (cf. Cribb & Gewirtz, 2007). The principals expressed, for example, that it would have been preferable if the municipality and the school organizer had come up with clear directives so that they had not had to take so many complex decisions on their own.

Another important result to consider is linked to the developmental domain, in which almost all systematic improvement work was paused to address the situation. Against this backdrop, it becomes essential to consider how improvement work might be ensured if there are recurring crises in society. It is worth emphasizing, in this regard, that several of the participating principals in the study had quite extensive experience of being principals, but they were nevertheless not able to work with school improvement for this rather long period of time. On the other hand, there was indeed development, due to the fact that strategies for distance education had to be developed, or teachers needed a much higher ICT proficiency. One principal called this ‘pandemic improvement’. We argue that it also shows the resilience of schools as professional organizations. Although there was a pandemic, principals and teachers made it work somehow. This also shows how development is in the DNA of schools as public organizations, beyond all efficiency discourses. Put differently, schools are dynamic organizations responding to societal changes.

Regarding the educational domain, we argue that the impact of the pandemic was somewhat restricted. COVID-19 did not overcome the rationale of school governance in Sweden. School quality, as well as the provision of special support to individual students, is in this context always related to output measures such as grading, and also national curriculum testing (Magnússon, 2015). As long as the student results were adequate, the principals had very extended autonomy, even during the COVID-19 crisis. The most significant measure of principals as frontline bureaucrats is student results. This is where their power is, and here we should dig further into practices of leadership in Lipsky’s (2010) traditions of street-level bureaucracy conceptualizations, in terms of information asymmetry or cherry picking strategies (see for the case of schools: Vedung, 2015). The principals in our study sometimes spoke about challenging boundaries in the best interests of students and teaching staff. In that sense, the findings echos the results of former work presented in the introduction (see e.g. Ahlström et al., 2020; Hooge & Pont, 2020; Jarvis & Mishra, 2020) showing the importance of school leader responsiveness and flexibility in unpredictable times including courage and commitment when facing complex situations, dilemmas and conflicting interests.

Accordingly, this explorative study both resonates and extends the research carried out on principals’ work during the COVID-19 pandemic. For instance, aside its wider contribution offering important insights on ‘pandemic leadership’ from a case that mainly held schools open, the study and its analytical devise also provide tools for understanding and analysing autonomy and control of principals’ work —

both in ‘normal’ circumstances and in times of crises. A key finding linked to the work of Cribb and Gewirtz (2007) is also that there is always a need to look into the relationship between autonomy and control and also to avoid normative assumptions, which becomes even more evident in times of crises. For example, instead of taking many complex decisions on their own, i.e. having a high degree of autonomy, principals instead preferred to receive clear directives from superintendents, local or national politicians, national school agencies.

Despite a robust analysis instrument and a careful sample of respondents, it is relevant to reflect upon the findings and some potential shortcomings of this exploratory study. To start with, there was a majority of male principals interviewees, which is notable given the ‘feminization’ of the principal profession in Sweden. However, the analysis found no evident examples of differences in terms of gender. Some variances were detected based on if principals worked in secondary or upper secondary school and/or if they worked in the same municipality, but overall, such differences were less evident in the analytical work. A variable that, on the other hand, could have a greater impact is the fact that all principals were rather experienced or sometimes very experienced, which reasonably affects both how they understand their autonomy in general, but also how they think and act in a pandemic crises. Accordingly, how more novice principals experienced and navigated their autonomy during this period and what support they (did not) receive might still be an unexplored issue.

Worthy of reflection is also that all principals worked in urban areas, which generally means that they had access to different types of supportive structures in their municipal organizations. This, indeed, becomes an important element to consider in terms of autonomy and control. Given that about 125 of Sweden’s 290 municipalities have a population of fewer than 10,000 inhabitants, and 25 of these have fewer than 5,000 inhabitants (The Swedish Association of Local Authorities and Regions, 2017), we know from former research that the conditions for principals’work can look very different in these regions (see e.g. Nordholm et al., 2021, 2022). Aginst this background, we conclude that there is a need for further studies, both qualitative and quantitative, to explore potential similarities and differences between various leadership contexts, within and between different countries.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The work was supported by the Norges Forskningsråd [315147].

Notes

1. The Swedish terminology for the authority is worth mentioning here, in order to illustrate the paradigm shift for an international audience. ‘The National Board of Education’ was called the Skolöverstyrelsen, translated literally, ‘The higher authority on schools’. Today, the Swedish National Agency for Education is called the Skolverket, meaning, again, literally translated, The agency for schools. Consequently, we see a shift from authoritive and bureaucratic leadership to governance.

2. Autonomy gaps are the difference between the amount of authority that district school principals think they need in order to be leaders and the amount they actually have (Adamowski et al., 2007).

3. We have elaborated on and tested the application of Ingersolls’ conceptualizations both for teachers and principals in the contexts of various European countries at different places (Wermke et al., 2023; Wermke et al., 2022).

4. In Sweden, work with student health is manifested in so-called student health teams (EHT, Elevhälsoteam). Regarding the Swedish Education Act, schools are obliged to have teams consisting of special educators, doctors, psychologists, social workers and the principal. The principal is always leading the group and has the legal responsibility for its decisions and recommendations.