Conference Agenda

This symposium aims to explore the autonomy of superintendents and school principals in Sweden, Finland, and Iceland from a comparative and historical perspective, with a particular focus on school improvement and inclusive education. This is significant, because formulating educational policies that can foster effective school improvement and a vibrant school leadership in schools has become an important task. Previous research has also revealed the importance of involving and taking into account all school levels (i.e., national, state, LEAs, and local school) and making them work in an integrated way (Hopkins, et al., 2014). These developments raise important issues on governance in more or less centralized or decentralized educational systems, with more or fewer stakeholders with more or less discretion. This also puts educational leadership autonomy into focus, i.e, the power of superintendents and principals to form their local processes in relation to the important questions of the 21^st century.

Sweden, Finland, and Iceland are relevant to compare because they are Nordic Model countries that draw on values such as universalism and local participation. However, there are also significant differences regarding the decentralization and accountability reforms of the last 30 years, with apparent impacts on the governance and work in school organizations (Dovemark et al. 2018). By comparing superintendents’ and principals’ experienced autonomy in municipalities and schools across the three countries, this symposium will deepen knowledge of the nature of educational leadership in relation to various forms of educational governance, and the tension between decentralization and centralization. We will show that, in this tension, various configurations of leadership autonomy can emerge as manifestations of different governance structures.

The symposium presents four studies that investigate the phenomenon of leadership autonomy from comparative and historical perspectives. We argue that this is of high significance because it expands the understanding of how educational organisations interrelate with particular nation-specific particularities. The aim of this symposium is not normative, even though it advocates certain autonomy principals and superintendents must have for the best possible kind of public education. Rather, the symposium aims to contribute to a further understanding of which configurations of leadership autonomy result in which kinds of educational organisations. For example, schools might reject autonomy in some dimensions of their work but desire greater discretion in others, in order to enact their educational leadership in a particular way.

The symposium’s contributions are characterised by various methodological approaches. We start with presenting the results of a systematic literature review of international research on leadership autonomy, also resulting in an empirical-based definition of leadership autonomy. Then we present the results of an interview study with 75 Swedish, Finnish and Icelandic principals and superintendents. We analyse the material by employing a multidimensional model that aims to render autonomy in public education empirically investigable. We will present a matrix that unites four different domains in which autonomy can evolve (educational, administrative, social, developmental domains). Contributions three and four will investigate leadership autonomy by focusing on Finnish and Swedish principals’ enactment of inclusive education policy, building on 60 interviews, and by investigating the autonomy of local leadership in different Swedish large scale school improvement projects since the 1980’s (drawing among others on extensive archive work)

This contribution maps and synthesizes the literature of school leader autonomy (SLA). Given international policy trends in which school systems have transformed from centralized to more decentralized, deregulated and marketized systems (see e.g. Ball, 2009; Whitty, 2010) the issue of SLA has evoked increased attention. That is, these changes generate interesting but also complex conditions for school leaders’ work in many contexts, e.g. by combining ideas on leadership autonomy with ideas on accountability and control mechanisms. Until now, no attempts have been made to systematically review the literature on SLA, and therefore we still lack a comprehensive and even critical image of SLA (cf. Cheng et al., 2016). In order to do so, we applied the PRISMA-model (see e.g. Moher et al, 2020) for conducting and reporting systematic reviews, generating 113 articles (n = 113) for final analysis. To attain rigor in the analytical work realist synthesis protocols were developed for each article. The research questions posed are of descriptive nature, addressing the what, who, where, and so what of the collected research corpus on SLA. Aside from descriptive details of the current body of research (e.g. distribution on publications based on country; distributions of publications over time; methodological approaches), this contribution reveals that some decisions by administrators and school leaders (e.g. on curriculum, instruction, administration, budgeting, organization) are addressed more frequently, while others (e.g. on values, democracy, racism, student well-being) have been more sparsely addressed. In research focusing on actors and their relations, principals are the most researched actors while few studies have focused explicitly on superintendents. A significant amount of research has instead approached SLA as a relational phenomenon that takes place between principals and other actors. Research has also addressed contextual particularities, but the number of articles that systematically consider the relationship between such particularities and SLA is limited. Furthermore, few studies employ statistical modeling allowing robust knowledge of the impact of increased SLA. Instead, most use school leaders’ perceptions. Based on the overall results, the contribution also provides a definition of SLA forming a valuable platform for future research. Altogether, this contribution demonstrates that SLA is a key concept for understanding the complex relations between education policy and education practice.

Ball, S.J. (2009). Privatizing education, privatizing education policy, privatizing educational research: Network governance and the ‘competition state’. Journal of Education Policy, 24(1), 83–99. Cheng, Y.C., Ko J., Lee, T.T.H. (2016). School autonomy, leadership and learning: A reconceptualization. International Journal of Educational Management, 30(2), 177–196. Moher D., Liberati A., Tetzlaff J., Altman D. G. (2010). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Medicine, 8(5), 336–341. Whitty G. (2010). Marketization and post-marketization in education. In Hargreaves A., Lieberman A., Fullan M., Hopkins D. (Eds.), Second international handbook of educational change (pp. 405–414). Springer.

In this contribution, we compare Leadership autonomy domains by investigating the decision-making of 47 Swedish, 32 Finnish and 17 Icelandic principals and superintendents. The theoretical point of departure in this contribution is Ingersoll’s (2003) theorizing on power distribution. His work also considers the role of school leaders. Decision-making abilities are related to the degree of conflict in the school environment, and autonomy is highly influential in issues related to daily activities. Accordingly, we have asked, (1) What are the most important decisions to be made in relation to school leaders’ work? (2) Who is allowed to make such decisions? The school leaders’ voices have been deductively analyzed using an analytical device to investigate autonomy in education from a comparative perspective. This device builds on the idea of autonomy in education as a multidimensional phenomenon (Ingersoll, 2003). This means that actors in school systems can obtain and/or lack different forms of autonomy in different dimensions of their profession (Wermke & Salokangas, 2021). Drawing on the work of Wermke et al. (2022), we propose four domains in which School leaders make decisions and can be controlled. Firstly, by the educational domain, we refer to matters related to activities and responsibilities connected to teaching and learning, including content, materials, and results. (b) In the social domain, grouping students is included or negotiating the rules of conduct, parent contact or conflict solution. (c) The Developmental domain refers to decisions that relate to issues of staff development or systematic school improvement. (d) Finally, by administrative domain, we refer to the administrative actions such as budget work, recruitment, admission etc. The most interesting results reveal, first, similarities between Nordic principals and superintendents, and, second, differences among principals within the countries. For principals, most decisions are made within the administrative domain, followed by the education domain. For superintendents, this distribution is the same, but the decision-making in the administrative domain is significantly bigger. From our domain-specific perspective on decision leadership autonomy, are superintendents mainly administrative leaders, while principals are both educational and administrative leaders? The comparative analysis leads us to the assumption, however, that the nature education leadership is, first of all, a complementary (and not an opposite) relation between administration and education. We call this educational leadership by administrative means. In reference to Hopmann's work, we call this bureau-pedagogy, noting that educational leaders must negotiate what is (educationally) legitimate and (administratively) legal.

Ingersoll, R. M. (2003). Who controls teachers’ work. Power and Accountability in America’s Schools. Harvard University. Wermke, W., & Salokangas, M. (2021). The Autonomy Paradox. Teachers´ perception on self-governance across Europe. Springer.

This contribution addresses the topic of policy enactment of inclusive education policy, as described by Finnish and Swedish principals. Both groups are seen as embedded in a ‘Nordic model of education’ (Larsen et al., 2021), supposedly sharing common values and aims manifested in terms of equity and universal education for all. The comparative approach focuses mainly on similarities in the both national contexts. Thereby searching for aspects that can be seen as generic for principals’ actions in the North. The study is based on interviews with 30 Finnish, and 30 Swedish principals. Our material was then analysed using qualitative content analysis (Mayring, 2007). In our findings, we observe major limitations of the principals’ possibilities to implement inclusive education. In order to operationalise inclusive education policies, principals in Finland and Sweden utilize a small number of strategies in what we label as the organizational toolbox. The tools available are i) bureaucratic rules and regulations, and ii) economic resources in terms of time and expertise to organise (special) education support, resources that are always scarce. Furthermore, we see found two distinct legitimation patterns justifying the principals’ actions, a) special education triage, and b) guardianship of the majority. These patterns show that there always are rudiments for decision making, thus making them non-arbitrary and of legal certainty. The special education triage means that the principals have to prioritise students’ different needs of special education support. The students’ needs are furthermore always related to a perceived normality and consideration of the majority of students. When using their professional agency, principals resort to practices of inclusive education through special education professions. This to maintain a balance between individual and collective needs. Our findings indicate the existence of a Nordic model of inclusive education which allocate considerable resources to include every student in meaningful education. However, when the needs of the majority of students are perceived as being in risk, collective needs are deemed more important than individual ones. Thus, resulting in exclusion strategies to preserve inclusion for the collective.

Larsen, J.E., Schulte, B. & Thue, F.W. (2021). Schoolteachers and the Nordic Model - Comparative and historicaL perspectives. London: Routledge. Mayring, P. (2007). Qualitative Inhaltsanalyse. Beltz.

This contribution examines national education agencies in Sweden and their changing organisation of large-scale school intervention between the 1980s and the 2020s. The material employed was identified through extensive archival research and a literature search. Drawing on Swedish reform history since the 1980s, we identified three periods of shifting governance regimes characterised by increased state efforts in large-scale school interventions. Three different state school agencies were in charge during these three periods. Drawing on the material, we present three context-specific models of state intervention. The models differ in their legitimation, funding, evaluation, and autonomy in the intervention provided to local leaders. Based on inductive analyses, we propose: the bureaucratic intervention model, building on ideas of a line of authority and local stakeholders as civil servants; the enforced cooperative intervention model, drawing on convincing local stakeholders about their apparent needs by guiding dialogic approaches and the standardisation intervention model, focusing on local stakeholders’ following of standardised improvement procedures and the achievement of education standards, controlled by authoritative means. The models reorganise the relations between state agencies and local stakeholders. While the first model aims at the decentralisation of education, the two other models’ ambition at re-centralisation with different intensities and stakes. We conclude that state agencies organise school improvement interventions in varying ways, with different consequences for local autonomy. We show that centralised governance regimes must not necessarily correlate with less local autonomy and vice versa. Even if the bureaucratic model is more centralised, it still enables local practitioners’ autonomy in school improvement, drawing on local expertise and capacity. In contrast, the standardised intervention model operates in a still decentralised system but leaves only restricted local autonomy. This finding confirms the argument that decentralised structures do not necessarily mean more local autonomy (Ingersoll, 2003). Given the Swedish “journey” from a centralised system to a decentralised one and back to a fairly centralised variety, this case offers insights valuable for various contexts. The design of large-scale school improvement interventions must always consider the logic of centralisation and decentralisation in school governance and their specific dynamics. Consequently, large-scale interventions are contingent. Moreover, their design always responds to context-specific conditions, challenges, and governance ambitions. Finally, school improvement interventions encompass processes of translation, negotiation, and sense-making. Such processes also involve the national agency level. The state agency level is seldom an actor that solely implements predetermined ideas into local school practice.

Ingersoll, R. M. (2003). Who controls teachers’ work. Power and Accountability in America’s Schools. Harvard University.