Understanding implicit reference societies in education policy

ABSTRACT

This study examines the reference societies of Norway and Sweden embedded in their education policy documents. We examined 4,260 bibliographic references in 19 white papers and green papers prepared for the 2016/2020 renewal of the Knowledge Promotion Reform in Norway and the 2015/2018 Knowledge Achievement Reform in Sweden. In addition, we interviewed 10 policy experts who participated in the preparation of the analyzed policy documents. The results show that the reference societies overall reflect the existing knowledge production and dissemination mechanisms in education policy; however, they significantly differed between Norway and Sweden regarding whether and to what extent they reference knowledge produced in other Nordic countries. Specifically, while Norway drew extensively on knowledge from its neighbors, particularly Sweden, Sweden seldom referenced knowledge produced in other Nordic countries. Policy actors identified similarity, relevance, accessibility, reform contexts, and institutional arrangements as reasons for (not) referencing neighbors. This study calls for further consideration of the political, social, and cultural embeddedness of the ‘socio-logic’ to understand implicit reference societies.

KEYWORDS:

• Reference societies
• education policy
• comparative education
• Norway
• Sweden

Introduction

National policymakers have increasingly made references to knowledge or information produced abroad as part of their domestic policy process. Such references include, among others, ranking tables and reports produced by international organizations and policies implemented in other countries. In comparative education policy studies, scholars have long tried to understand why and when policy actors make these kinds of international references. What their studies have found is that policy actors often externalize when they need to form alliances or seek greater legitimacy to pursue their political agenda. They have also found that despite the development of global education policies, national political contexts and the institutionalized policy process in each country seem to play a crucial role for the ways in which international references are used at the local level (Baek 2022; Schriewer 1990; Schriewer and Martinez 2004; Steiner-Khamsi 2004). However, not all countries receive the same level of interest and attention in this practice of borrowing from external educational systems. Instead, each country seems to have its own ‘reference societies,’ from which they borrow policies and practices or to which they refer either positively or negatively in the policy process (Bendix 1978; Waldow 2017). Such reference societies not only include other countries that are considered the focal country’s competitors or threats but also collaborators and partners. The choice of reference societies is based on various contextual factors, ranging from political and cultural similarities to geographical proximity (Lingard 2011; Waldow, Takayama, and Sung 2014). Furthermore, new reference societies are constructed based on Organisation for Economic Co-operation and Development (OECD) membership and successful performance in international large-scale assessments (ILSAs) such as programme for international student assessment (PISA) (Baek 2022; Santos & Centeno, 2023; Sellar and Lingard 2013).

Analyzing reference societies in national educational policymaking in the Nordic region is of special interest, because it has been perceived that the Nordic countries serve as reference societies for each other within the region, perhaps due to their historical, cultural, and political similarities as well as the rich communication and cooperation among the Nordic countries in education policy (e.g. the Nordic school meetings beginning in 1860; Landahl 2015; Tröhler 2023). The idea of Nordic policy learning has been reinforced by the establishment of forums such as the Nordic Council of Ministers. Scholars have also discussed that consensus on policy priorities and challenges (Papakosma 2023) as well as consensus in values and beliefs (Nedergaard and Nobel 2022) facilitate and contribute to the use, exchange, and transfer of policy knowledge, and that the ‘consensus’ is perceived to be salient among Nordic countries (Nedergaard and Nobel 2022).

Despite these common perceptions on similarities and shared values, few empirical investigations have been conducted regarding which of the Nordic countries are used as reference societies and to what extent, and why, policy actors actually make references to other Nordic countries in education policymaking. Some scholars have investigated how and why policy actors in Norway and Sweden make or avoid making international references, although their focus was not specifically on references to other Nordic countries (Baek et al. 2018; Grek 2020; Pettersson, Prøitz, and Forsberg 2017; Prøitz 2015). Waldow (2009), for example, investigated Sweden’s post-WWII policy transfer and coined the term ‘silent borrowing’ to describe its policy borrowing without any explicit reference to other systems. He argued that Sweden’s self-perception of the global supremacy of its education system discouraged policy actors from clearly indicating any borrowing or lesson-drawing from other countries. Moreover, making international references was not an effective strategy to gain legitimacy in the Swedish political context at the time. However, a more recent study has observed that in recent years, Sweden has started to use ‘international argument’ as a prominent strategy to justify policy changes (Nordin and Wahlström 2022; Ringarp and Waldow 2016).

By contrast, little work has explored Norway’s reference societies in education policy. Many studies have examined international influence and policy borrowing in Norwegian education policy, but without identifying particular countries as reference societies. Among the few exceptions, Sivesind (2019) examined how Norwegian policymakers perceived Finland as a country for emulation in school reforms; however, the study’s analysis of reference societies was limited to Finland.

This study aims to further explore the complex and sometimes unpredictable reference societies that are cited in national policy documents. Norway and Sweden serve as particularly interesting cases for comparison because of the similarities in their institutionalized policymaking processes as well as their geographical and cultural proximity (see Steiner-Khamsi et al. 2022). We examine bibliographic references in the policy documents prepared for the most recent school reforms in the two countries, respectively, and interview central actors who served on national ad-hoc commissions. Thus, this study provides insights into both the structures of references in the official policy documents and the behind-the-scenes process, where various factors and rationales influence what eventually becomes an official reference. This study does not intend to address educational reforms or policy processes in a broader sense; instead, it specifically focuses on the use of policy knowledge originated from external systems in educational policymaking.

Understanding references in education policy as the production of legitimacy

This study takes its theoretical point of departure in the literature on policy borrowing and lending (Steiner-Khamsi 2004). Here, attention is drawn to the role of local policy context in the examination of why and to what extent reforms, international standards, and/or best practices are adopted at the local level. Directing analytical focus towards the role of local policy contexts means taking seriously aspects such as agency, impact, and timing in processes of policy borrowing.

Instead of interpreting policy borrowing and lending as simply a rational process of exchanging best practices, this tradition acknowledges that borrowing occurs for different political, cultural, and economic reasons. In this study, we specifically focus on the political dimension, which underscores how policy borrowing is used as a ‘certification’ strategy to validate and legitimize political actors and their performances during the transformative period (Steiner-Khamsi, 2010). Legitimacy in this political dimension is not a quality inherent in certain policy elements, but rather the result of an interactive process where legitimacy is being produced (Waldow 2012). The production of legitimacy occurs when policy agendas are seen as desirable and useful within the local arena. A prominent strategy among local policy actors to gain such legitimacy, which has garnered significant attention from researchers, is the ‘externalization thesis.’ Originated from Luhmanns’ sociological systems theory, the concept addresses systems using external points of reference. These points of reference are then recontextualized and adapted within the logic and language of the new context. Inspired by Luhmanns’ work, Schriewer (1990) has introduced the concept of ‘externalization to world situations’ to explain how internal policy reforms are legitimized through external referencing such as international trends or examples.

In this study, we employ Schriewer’s concept to understand the use of the external references in policymaking in Norway and Sweden. However, externalization to world situations is only one facet of externalization in education policymaking. Another critical type, of particular relevant to this study, is what Waldow (2012) refers to as ‘externalization to principles and results of science.’ This occurs when legitimacy is produced through arguments that educational reforms are firmly grounded in scientific evidence. Bringing together the concepts of ‘externalization to world situations’ and ‘externalization to principles and results of science’ facilitates a comprehensive discussion on the use of external references in education policymaking. Within the broader context of evidence-based policymaking, references serve to validate and provide legitimacy to the evidence presented in policy documents. Steiner-Khamsi (2022, 35, 39) writes:

References help validate or provide legitimacy to the evidence that the author (e.g., the government-appointed expert commissions or the government) has presented in the document. Thus, if ‘evidence is assertion backed by information’ (Cairney 2015), then a reference is validation of evidence. Said differently, references are used to provide authoritative status to the evidence presented in policy documents. […] [I]n comparative policy studies, the term ‘reference’ also carries a spatial, geo-political, or epistemological connotation.

This highlights how references can legitimize policy ideas through externalization to external educational systems, further supporting evidence-based policymaking (‘externalization to world situations’ and ‘externalization to principles and results of science’).

Moreover, the analytical approach of this study involves understanding externalization as a selective process. Not all countries, international organizations, or types of knowledge hold equal interest for policymakers at any given time. Not all of them become catalysts in the local production of legitimacy. However, those that are found attractive, relevant, and useful to policy actors may serve as a reference society. In this study, the concept of ‘reference society’ opens up for a discussion about where Norway and Sweden draw upon policy knowledge in their respective national production of legitimacy and to what extent they contribute to each other’s legitimization.

Research on reference societies

The concept of reference societies was coined by Reinhard Bendix (1978) in his book Kings or People, where he applied the concept broadly to policy actors’ reactions to ‘the values and institutions of another country’ (p. 292, as cited in Waldow 2017, 648). In practice, however, the application has been somewhat more limited to understanding the ‘model nation from which to borrow elements’ (Waldow 2017, 647). In comparative education, the concept was first applied by Butts (1973), who observed that the governments in developing countries frequently referenced the education systems in the Global North as a model for emulation (Steiner-Khamsi 2022; Waldow 2019). Conversely, in his work on reference societies in the German education policymaking debate, Florian Waldow (2017) points out that while the traditional application in comparative education often highlights positive references to another country, Bendix’s original definition encompasses both positive and negative references (as a model for emulation or counter-reference). Furthermore, in recent years, scholars have called for expanding the country-specific concept of reference societies to include reactions to sub- and supra-national regions (Santos and Centeno 2023). For example, East Asian and Scandinavian education systems have become common reference societies for many countries.

The growth and spread of ILSAs have contributed to the frequent use of reference societies in education policy debates. ILSAs allow national policymakers to assess the success of other national systems based on their ILSA performance and to refer to them in policy debates for glorification, scandalization, or problematization (Steiner-Khamsi 2004). For example, the education systems of Finland, Singapore, Shanghai,¹ and South Korea have served as reference societies for many countries due to their high ILSA performance, although they have met with both positive and negative receptions. Furthermore, OECD has emerged as an attractive reference society, accompanied by the increasing influence of its ILSA – PISA – as well as other governing instruments.

Empirically, previous scholarship has examined reference societies by considering the explicit and direct references to foreign countries or systems in policy documents, speeches, and media articles. However, this empirical approach does not allow researchers to identify the implicit reference societies on which policy actors draw in the policy process. In particular, in an era of evidence-based policymaking, it is important to understand what kinds of evidence are used in the policy process and where they are produced. Scholars, for example, have pointed out that much knowledge platforms and publication venues are located in the United States and the United Kingdom.

To this end, we argue that studies on reference societies need to encompass policy actors’ reactions to ideas and knowledge produced in other countries. Indeed, previous studies have drawn on the bibliometric data of research journals to examine reference societies in ‘academic and scientific knowledge’ (e.g. Schriewer and Martinez 2004). In recent years, bibliometric data of policy documents have also been used to investigate reference societies in policy knowledge. This study is part of the international research project, Policy Knowledge and Lesson Drawing in Nordic School Reform in an Era of International Comparison (POLNET),² which also examines bibliometric references in the policy documents to compare the evidence use and transfer in education policy in the five Nordic countries (see Karseth, Sivesind, and Steiner-Khamsi 2022, for further details). In this study, we examine the countries from which Norwegian and Swedish policy actors draw evidence by identifying the publication location of references used in policy documents. We follow Bendix in his open use of the concept of reference societies, which enables us to explore how international and regional knowledge are used as external points of reference in the production of legitimacy in national education policymaking.

Reform contexts

Some information on the reforms in Norway and Sweden can help to contextualize reference use in the policy process. The reforms included in this study are incremental reforms following prior fundamental reforms. In the following, we present the main constituents of the two reforms, here referred to as the Norwegian School Reform 2016/2020 and the Swedish School Reform 2015/2018. In relation to the Norwegian reform, there are two white papers titled St.meld.nr. 28 (2015–2016): Fag – Fordypning—Forståelse – En fornyelse av Kunnskapsløftet [Report No. 28 to the Parliament: Subjects, In-Depth Learning – Understanding. A Renewal of the Knowledge Promotion Reform] and St.meld.nr. 21 (2016–2017): Lærelyst – tidlig innsats og kvalitet i skolen [Report No. 21 to the Parliament: Eager to Learn – Early Intervention and Quality in Schools]. In the Swedish case, there is one white paper titled Prop. 2017/18:182 Samling för skolan [White Paper 2017/18:182 Gathering for School].

The Norwegian School Reform 2016/2020

The Norwegian School Reform 2016/2020 follows the fundamental Knowledge Promotion Reform launched in 2006. The 2006 reform, which brought about increased decentralization and test-based accountability (see Steiner-Khamsi et al. 2022), caused a radical shift in Norwegian schooling, replacing input-oriented policy instruments with output-oriented ones such as data-based planning, measurable objectives, and standardized tests (Møller and Skedsmo 2013). With the aim of fostering ‘a culture of learning’ (Karseth and Sivesind 2010, 109), the reform promoted the development of general competencies among pupils instead of knowledge of specific school subjects. National tests were made public, which simultaneously made it possible for stakeholders to rank and compare schools. In 2013, the Ministry of Education and Research stated that the reform could be considered a success, and the PISA results began to increase. However, the Ministry of Education and Research also stated that time had revealed several shortcomings of the reform, resulting in an overloaded curriculum and a need to better specify the key elements to be learned. Enhancements in social democracy, citizenship, sustainable development, public health, and well-being were highlighted as areas of special importance for the future of Norwegian schooling, along with special investments to support low-performing pupils from grades one to four in reading, writing, and numeracy.

To address the identified shortcomings, the Ministry of Education and Research initiated two parallel reform processes: Renewal of the Norwegian Knowledge Promotion Reform (Ministry of Education and Research 2016) and Early Intervention and Quality Monitoring (Ministry of Education and Research 2017). These two white papers make up what we here refer to as the Norwegian School Reform 2016/2020, as the first white paper was published in 2016 and the reform was launched in 2020.

The Swedish School Reform 2015/2018

In many ways, the Swedish School Reform 2015/2018 can be viewed as a supplement to the previous fundamental reform of 2011, through which Swedish politicians sought to address what was then described as a national school crisis due to continuously declining results on the PISA test (Nordin 2019). The 2011 reform actually involved the launch of a wide range of complementary reforms intended to achieve stronger state control over compulsory schooling. Critics argued that the decentralization of the Swedish school system in 1991, whereby the 290 municipalities took over the mandatorship from the state, had resulted in excessive variation between schools across the country in terms of both teaching and performance. To make national schools and schooling more equivalent and raise the level of knowledge among pupils, the government launched a new school law, a new national curriculum for compulsory schooling, a new grading scale, and new national tests together with a new national teacher education. However, despite the extensive reform package, the PISA results continued to drop, reaching a historical low in 2012 and exacerbating the concerns of already uncertain politicians. At this difficult moment, the Swedish government turned to the OECD for help (Nordin and Wahlström 2022; Pettersson, Prøitz, and Forsberg 2017).

The initiative meant that the OECD initiated a period of study visits and analyses resulting in the report Improving Schools in Sweden: An OECD Perspective (OECD, 2015). In the report, the OECD urged the Swedish government to increase its efforts to meet the country’s current challenges, highlighting three areas of special importance for Sweden to fulfill its commitment to excellence and equity: 1) strengthening quality through an increased focus on equity, 2) building a high-quality teaching profession, and 3) steering policy and accountability towards improvement.

In response, the Swedish government appointed a commission called the 2015 School Commission [2015-års Skolkommission], whose work resulted in a green paper titled Gathering for School – A National Strategy for Knowledge and Equivalence (SOU 2017:35). The OECD report and this green paper thus became central providers of expert knowledge for the government in writing the white paper with the same title Gathering for School (Prop. 2017/18:182). The white paper effectuated the incremental Swedish School Reform 2015/2018 (where 2015 refers to the launch of the OECD report and 2018 to the launch of the white paper).

Research design

This study examines the bibliographic references in the policy documents used to prepare the most recent school reforms. As demonstrated in the previous section, in both Norway and Sweden, the Ministry of Education and Research prepared a white paper (‘Stortings melding’ in Norway and ‘Proposition’ in Sweden) when proposing a new reform. These white papers were supposedly based on the green papers from government-appointed advisory commissions (‘Norges offentlige utredninger’ (NOU) in Norway and ‘Statens offentliga utredningar’ (SOU) in Sweden). We first identified three white papers that contributed to the development of the Norwegian School Reform 2016/2020 and the Swedish School Reform 2015/2018. We then selected the green papers of the advisory commissions that were explicitly cited in the white papers. The selected policy documents constitute ‘official policy knowledge,’ because they were authored by experts who were commissioned by the government, and these sources were acknowledged as foundational references for the government’s policy proposals. We then extracted the bibliographic references from the 19 source documents. The references can be used to either support or disagree with specific ideas. However, we do not distinguish between the nature of these references in this study since both usages serve the functions of legitimacy and validation. In total, our sample includes 19 policy documents and 4,260 references.³ The average number of references per document for Norway and Sweden was 265 and 179, respectively.

Instead of counting direct and in-text references to foreign countries in policy documents, this study looks at the reference list to identify the knowledge base of the policy documents and then coded the publication location of the 4,260 references to identify the reference societies of Norway and Sweden. This expands the previous conceptualization of reference societies from direct referencing of an external system to implicit referencing mediated by knowledge produced in an external system. Due to the presence of multinational publishing companies and international organizations, references to knowledge published in particular countries may not be directly related to the knowledge produced in those countries. However, we argue that this reflects current knowledge production and dissemination mechanism in education knowledge, policy, and practice. For the most prevalent reference societies, we also looked into understand which documents are published in the countries (e.g. OECD documents published in France).

After the bibliometric analysis, we conducted interviews with five policy actors in Norway and five in Sweden in March – April 2022. Interview requests were sent to the list of commission members who had participated in the policy process of the two education reforms of interest and whose contact information could be retrieved. We received confirmation from total 10 commission members including government officers from the Ministry of Education and Research, academics, and non-government organization members.⁴ The interviews provided additional insights on the patterns of reference societies discovered in the bibliometric analysis.

The interview questions addressed a range of topics, including the sources of policy knowledge used in discussions and meetings related to the white papers and green papers. Additionally, we inquired about the methods of communication and knowledge exchange among members. We also asked questions specific to reference societies, addressing their utilization of knowledge from other Nordic countries in education policymaking and their engagement with knowledge and information beyond the Nordic region to inform the policy processes. All interview recordings were transcribed. We first conducted the open reading of the interview transcripts and then identified emerging codes and categorized them by themes that address research questions. Throughout the analytic process, two researchers closely communicated to review, compare, and refine the themes.

Findings

Comparison of reference societies

We identified the reference societies of Norway and Sweden by examining the publisher locations for the references in policy documents (Figure 1). The results show that Norway’s top five reference societies are the United States, France, the United Kingdom, Sweden, and Denmark. Upon closer examination, the references from the United States and the United Kingdom mainly arise from international academic publishing companies based in those countries, such as Cambridge University Press, Routledge, SAGE Publishing, and Blackwell, and intergovernmental organizations such as the United Nations, headquartered in New York. The references from France are mostly OECD documents published in Paris, and about 39% of Swedish references are reports published by the Swedish National Agency for Education (Skolverket). Sweden’s top three reference societies are the same as Norway’s (the United States, the United Kingdom, and France) and for the same reasons, that is, publishing companies and intergovernmental organizations constituted a large portion of the references from the United States and the United Kingdom. The other two countries in Sweden’s top five are Belgium and Canada. The references from Belgium are primarily due to the publications from the European Commission and the European Parliament, and the references from Canada are mostly drawn from Ontario.

Figure 1. Publisher locations by country for references in Norwegian and Swedish policy documents.

Note. The figure includes only the countries where more than three references of either Norwegian or Swedish policy documents were published.

Despite the similarities in reference societies between Norway and Sweden, there is an interesting difference: Norway frequently draws on knowledge from its neighbors, largely from Sweden, whereas Sweden rarely references knowledge produced in other Nordic countries. Indeed, the number of references to Denmark (67), Sweden (74), and Finland (16) in Norwegian policy documents is much greater than the number of references to Denmark (7), Norway (5), and Finland (4) in Swedish policy documents.

Why do policymakers (not) reference other Nordic countries?

Interviews with policy actors allowed us to explore further why policy actors in Norway and Sweden use knowledge produced in the Nordic region differently. The frequent use of Nordic references in Norway can be attributed to factors such as similarity, relevance, and accessibility. Conversely, the lack of use, as explained mainly by Swedish policy actors, can be attributed to the differences in committee membership, reform contexts, and institutional arrangements.

Similarity and relevance

Policy actors in Norway described the use of knowledge produced in other Nordic countries as ‘quite natural.’ They consider this a common practice in the policy process in Norway because there is ‘an image that [Norway, Sweden, and Denmark] have similarities.’ Another interviewee described the Nordic countries as ‘a small unit’ and elaborated further on the similarities:

[Not only] geography and history but I think also we are quite similar when it comes to politics [and] ideology, so we have a great similarity, at least between Denmark, Norway, and Sweden, because there has been so much contact and collaboration and also some wars … but that’s a long time ago, so, I think perhaps first and foremost it’s not about geography or history … it’s about that we, in terms of political profile and culture, have a lot in common.

This observation suggests that the similarities in political profile and culture perceived by interviewees have influenced the selection of reference societies. Policy actors also added that they draw on knowledge from other Nordic countries because their education systems and concepts are commonly believed to be highly similar and relevant. For example, one interviewee referred to the concept of Bildung as an example of the Nordic countries’ shared educational ideas and concepts. The concept of Bildung, which stems from German philosophy, suggests that the purpose of schooling is not merely the transfer of knowledge from society to individual learners nor the transformation of scientific knowledge into the classroom. Instead, the Bildung tradition places greater emphasis on ‘the use of knowledge as a transformative tool of unfolding the learners’ individuality and sociability’ (Hopmann 2007, 115). Overall, Norwegian policy actors highlighted the difficulty of applying knowledge from different foreign countries and identified similarity and relevance as the main reason for drawing on knowledge from other Nordic countries.

I think generally we find Nordic countries of interest because of the similarities of context, and when you take a look, for example, at national reports from some international studies … you will see that the comparison, in-depth comparison is … 90% [with] Nordic countries. You can take a look at some of our national reports because those countries are of interest to us … we are interested in what we can learn from similar context[s], and it’s sometimes not so easy to use research results from countries. Both the society and educational systems are so different. So we know that Nordic [countries] are most relevant for us, and of course every time we use research we have to make our judgements [about whether] it is really relevant.

Several Swedish interviewees also referred to well-established networks and collaborations that lead to a greater level of similarity and relevance of education systems, concepts, and knowledge within the Nordic region. However, unlike their Norwegian counterparts, these observations are not reflected in the bibliographic references of the green papers. Despite the many perceived similarities between the Nordic countries and the benefit of similar languages between Sweden, Norway, and Denmark, many Swedish respondents said that there has been little or no formal exchange with these countries. Moreover, although not evident in the reference lists, Sweden seems to have a greater interest in Finland than Norway despite the language barrier because Finland’s strong performance in PISA lends great legitimacy to policy actors in policymaking.⁵ One Swedish policy actor said that even though their commission consulted Norwegian colleagues during the work, the insights gained were not reflected in the references or examples used in the green paper. Regarding examples and references, many Swedish interviewees shared that they followed the same logic as in their role as researchers, where they used international publications such as OECD publications or academic articles published in English as their authoritative sources.

[Referring to international publications] is a trend which characterizes research today, but [it also] spills over into the work of government appointed commissions, and Nordic comparisons are not necessarily taken into account in the same way as they were before.

Despite the many similarities with the Nordic countries, Swedish chairs and committee members prefer to borrow examples from the international policy arena and build on international knowledge in line with current academic trends rather than refer to research, or use examples, from other Nordic countries.

Accessibility

Accessibility of knowledge was another major reason for utilizing knowledge produced in other Nordic countries. In particular, accessibility of knowledge is closely related to similarity in language and familiarity with the policy systems. Norwegian policy actors shared that they felt comfortable communicating with Swedish and Danish policy actors and accessing policy documents or research findings published in Swedish and Danish. Indeed, Norwegian commissions include members from Sweden and Denmark, and the interview participants shared that although the meetings were conducted in Norwegian, the experts from Sweden and Denmark could participate in discussion without difficulty. By contrast, the interviewees shared that even among knowledge produced in Nordic countries that produced in Finland may be less sought after in the Norwegian policy process because of the language difference. This is consistent with the findings shown in Figure 1, as references to knowledge produced in Sweden and Denmark are much more frequent than those to knowledge produced in Finland.

In addition, another reason Norwegian policy actors gave for seeking knowledge produced in other Nordic countries is that ‘[policy actors] know where to find [knowledge].’ Due to frequent and close collaboration between the Nordic countries, policy actors have become familiar with institutions, actors, and platforms for education knowledge, which allows them to easily identify a potentially helpful source of knowledge when needed. For example, Skolverket—the Swedish National Agency for Education, which is equivalent to the Norwegian Directorate for Education and Training – is one of the most popular knowledge sources, constituting approximately 39% of the Swedish references in Norwegian policy documents. Skolverket is known to and perceived by both Swedish and Norwegian policy actors as credible and legitimate source of educational knowledge on its education system.

Membership composition

Another possible explanation for the (non-)use of Nordic knowledge utilization is the membership composition of government-appointed advisory commissions. The informants shared that the knowledge base of commissions was often heavily influenced by the knowledge base of the individual members. Thus, when there were members from Sweden and Denmark on a commission, there was a greater opportunity to learn from and seek knowledge produced in those countries. Indeed, commissions that included any members from other Nordic countries showed more frequent use of Nordic knowledge.

Even among the Nordic countries that are perceived to have a similar political profile and educational system, whether a commission includes a member from a particular country shaped how commission members used knowledge produced in that country. A Norwegian policy actor recalled:

In the last green paper, we had a committee member from Denmark … I think maybe we used Denmark more actively in our work than Sweden … I can remember we had some reflections about the Swedish system also. But Denmark was important for our committee work because one of the members came from Denmark.

However, fewer Finnish than Swedish and Danish experts served on the Norwegian advisory commissions, likely due to the language difference explained above. By contrast, Swedish policy actors turn to Finland for best practices more often than to other Nordic countries due to Finland’s PISA success. A Swedish policy actor shared that:

They [commissions] follow trends, and the trend here [in Sweden] is that Finland is highlighted. We also had a Finnish member in our commission … he had great influence, and I guess he was there because Finland was supposed to be a good example to compare with.

Thus, irrespective of reasons for participation, the composition of commissions is important for the commissions’ work and the references used.

The composition of the commission has become even more critical as commissions in Sweden have increasingly transformed from larger commissions to one-person commissions with additional members and experts linked to them (e.g. a small secretariat, a group of experts who are bureaucrats from the government departments, and a reference group consisting of academic experts). For the secretariat, and to some extent also other positions, the person leading the one-person commission has an important responsibility to select and hire people they find suitable for the task. In their daily work, the person leading these smaller commissions can decide what knowledge to include or not include. Both the composition of larger commissions and the role of the person leading one-person commissions imply that there is also always a subjective aspect to the understanding of knowledge utilization in Nordic policymaking. Although commissions often carry out research interviews, the commission can decide on whether and how to use the findings of the interviews. One of the Swedish interviewees shared that the background of the commission members has a tremendous impact on how the green paper is written, the number and type of references used, and the extent to which the chair writes the text personally or delegates parts of the writing process to members of the secretariat. Another interviewee said that when writing the green paper, they used references they knew beforehand and found valuable. For example, we found in an earlier study that for SOU 2017:35, the chair of the commission was one of the most-cited academic references (see Steiner-Khamsi et al. 2022, for further discussion). In many ways, commissions function as knowledge filters, determining what knowledge is finally used as evidence to support the proposals presented at the political level.

In addition, when examining the composition of the Swedish commission members, excluding secretariats, about 80% were civil servants and political actors, 10.53% were academics, and 8.42% were interest group members. In Norway, by contrast, about 26% of experts were interest group members, 22% were civil servants and political actors, and 34% were academics. This may explain the variation in the use of Nordic references. Civil servants and political actors are expected to, and are more inclined to, draw on knowledge produced within their own country, such as previous reforms and legal documents. The higher representation of civil servants and political actors in Swedish commissions, compared to the more diverse representation in Norwegian commissions, may have influenced how they utilize international and Nordic references in the policy process. Indeed, Sweden’s utilization of international references and Nordic references (18.93% and 1.27%, respectively) was lower than that of Norway (26.08% and 7.09%, respectively).

Reform context

Because Norway and Sweden demonstrate different patterns in Nordic knowledge utilization, it is also important to consider the knowledge exchange between Norway and Sweden. While Sweden is one of the main knowledge sources in Norwegian policy documents, Norway does not appear to be one of the major sources in Swedish policy documents. An informant shared that this may be due to the characteristics of the Norwegian School Reform 2016/2020. One of the main focuses of this reform was competency-based education, and Sweden was considered an early adopter of this approach. In the 1990s, Sweden introduced competency-based education as part of a Bildung-oriented curriculum reform, for which it experienced significant criticisms in subsequent decades. Sweden’s earlier adoption of the concept which the Norwegian School Reform 2016/2020 used may explain the Norwegian commission members’ interest in knowledge produced in Sweden. A Swedish policy actor shared:

We had a professor of education, Ulf P. Lundgren, who was the chair of the CERI commission in OECD, who worked with the competency-based curriculum at that time, and we actually imported a lot of those ideas into the Swedish curriculum reforms of the 1990s. So Sweden was a really early bird when it came to adopting competency-based education, and the reforms of the 1990s were really to a large extent building on those ideas and experiences from that work … . Norway did import some of those ideas in the ‘90s but not at all to the extent as in the Swedish case, but later on in the first decade of the 2000s, the public discussion on competency-based curriculum grew in Norway. At the same time, you could say that we had a backlash on competency-based education in Sweden … . So they [policy actors in Norway] were really interested learning from [Swedish] reforms in the 90s but also to learn from the critical discussions later on to avoid the [same] critical discussion in Norway.

Furthermore, the informant pointed out that historically, there has been extensive knowledge exchange and collaboration between researchers and policymakers in Norway and Sweden in the field of curriculum studies. Another Swedish interviewee confirmed that there is much academic cooperation within the Nordic academic milieu, especially between Sweden and Norway, even if it is not reflected in their references.

In Sweden, the OECD has become an integral part of the domestic policy discourse on education and one of the most important contextual factors that shape national policy actors’ priorities. One interviewee shared that Sweden’s declining results on PISA led to the initial appointment of their one-person commission. They also noted that the commissions undertook trips to Finland to visit schools, universities, and government agencies and that Finland, due to the country’s PISA success, was considered the obvious reference point for the commission’s work.

Institutionalized norms and arrangements

The interviews further revealed that knowledge use, and more specifically reference use, may not always be intentional or reflective. Not all interviewees could provide logical explanations for their use of particular references in the policy documents. One interviewee even stated that policy actors often ‘really have no idea.’ However, this lack of awareness does not mean that the knowledge selection is random; rather, we argue that it is heavily influenced by the institutional structure, norms, and culture to the extent that certain practices become implicit and unreflected. For example, interview data suggest that referring to international references from academic journals published in English is a habitual practice for many commission members in both Norway and Sweden.

Furthermore, differences in Nordic knowledge utilization between Norway and Sweden can be contextualized by the differences in the ‘expertise-seeking arrangements’ between these countries (Baek 2020). Although ad-hoc advisory commissions have served as the primary expertise-seeking arrangements across the Nordic region, some notable changes have occurred in the past decades. For example, while Sweden has increasingly replaced large advisory commissions with one-person commissions since the 1990s, as noted earlier, Norway has continued to seek knowledge through large advisory commissions. In practice, one-person commissions mean one person together with one or two secretaries, accompanied by an expert group consisting of bureaucrats and a reference group of researchers. Identifying relevant knowledge thus becomes a more individual process in one-person commissions, whereas the more collective process in large commissions increases the chance of including foreign knowledge and experts in the process.

Discussion

This study empirically examined the reference societies of Norway and Sweden embedded in the reference list of education policy documents. There are two overall observations that could be discussed within the broader discussion of knowledge production and use in education policy. First, the findings demonstrate that particular countries hold greater importance in transnational, geopolitical space of educational knowledge, echoing ‘the global economy of knowledge’ criticized by many scholars in international education (e.g. Takayama, Sriprakash, and Connell 2017). The referencing patterns suggest the United States and the United Kingdom remain the main sources of knowledge, as they host many multinational publishing companies such as Routledge, SAGE, Blackwell, and Springer. Furthermore, France, as the home of the OECD headquarters, has become a powerful and influential sociopolitical space for many education policy actors.

Another trend that needs greater attention to is an increased emphasis on references to international references in general. This study includes two incremental reforms, and it is often expected that incremental reforms are more likely to refer to internal experience with the previous, fundamental reform, that may lead to the greater presence of domestic references in the report compared to international references. However, a previous study in Norwegian context (Steiner-Khamsi et al. 2020) comparing the references of the policy documents between the 2006 reform and the 2016/2020 reform found that that is not necessarily the case. While 20.55% of references in the relevant white and green papers prepared for the 2006 reform were international documents, 27.17% of the references in the white and green papers prepared for the 2016/2020 reform were published outside Norway and the Nordic region.

A particularly interesting finding of this study is the significant difference between the reference societies of Norway and Sweden, regarding whether and to what extent they reference knowledge produced in other Nordic countries. Despite the countries’ close political and policymaking collaborations and connections, the overall use of regional references within the Nordic region has been limited (Volmari, Sivesind, and Jónasson 2022). Volmari et al. (2022) explain that Nordic cooperation and knowledge are natural and implicit parts of the policy process in Nordic countries to the extent that they are not explicitly referenced. This finding calls us to consider further how externally produced knowledge (e.g. Nordic or international policy and practice) becomes internal knowledge and when policy actors use external or internal knowledge for legitimation. When seeking legitimacy, while Norwegian policy actors prioritized knowledge produced in contexts that had greater perceived similarity and relevance to the Norwegian contexts and could be easily accessed, Swedish actors prioritized knowledge produced in contexts that hold greater importance in current geopolitical space of educational knowledge.

Indeed, Sweden’s lack of references to knowledge produced in other Nordic countries highlighted in this study, compared to Norway’s frequent references to knowledge produced in its neighbors, calls for further discussion related to previous literature. Drawing on the examples of the Swedish green papers from the early 1960s and early 1970s, Waldow (2009) points out that education policymaking in Sweden was heavily influenced by international contexts, but these were not explicitly acknowledged by policymakers and educational researchers, as explicit reference to external contexts was not considered an effective legitimation strategy within the Swedish political culture. Thus, the transferred policy and practice were instead presented as rational and scientific solutions. Furthermore, Sweden’s self-perception as a global pioneer in education shaped Swedish policy actors’ decisions regarding silent borrowing. In later studies, scholars have observed that ‘international argument’ has become a prominent legitimation strategy in Swedish political culture in recent years in response to Sweden’s disappointing PISA results (Nordin and Wahlström 2022; Ringarp and Waldow 2016). It has also been noted that in an era of evidence-based policymaking, the act of externalization (i.e. referring to international knowledge) itself has been internalized and has become part of the ‘rational’ policymaking process (see Baek 2022, for further discussion). The findings of this study confirm the discussions of previous studies. In addition, this study suggests that Sweden’s silent borrowing and self-image as a pioneer are still applied 1) within the Nordic discourses, where Sweden appears to have historically ‘occupied a hegemonic position’ (Andersson and Hilson 2009, 223), and (2) to a specific reform topic (i.e. competency-based education). Perhaps, the question is not only limited to the content of knowledge itself. A factor that may shape policy actors’ reference patterns could also include where the knowledge is produced and who produced the knowledge.

Furthermore, this study calls for the consideration of the implicit and unreflected. Studies on reference societies have focused on understanding the ‘socio-logic’ of a system that shapes the references to particular external systems (Schriewer and Martinez 2004). When investigating the socio-logic, recent studies have mainly focused on policy actors’ rational selection based on political calculations and (de)motivations for lesson drawing within the system. This study also found that policy actors used references based on similarity, relevance, accessibility, membership composition, and reform contexts. However, our study also suggests that not every reference or lack of reference can be explained or justified, because some may be due to institutionalized practice and coincidence. The construction of reference societies is not only based on (bounded) rationality but also shaped by the implicit and unreflected as it has been framed by history, institutional culture, and social norms. It is therefore particularly necessary to consider the political, social, and cultural embeddedness of the ‘socio-logic’ to understand implicit reference societies that contribute to the knowledge base of education reforms. This study contributes to existing scholarship on reference societies by expanding the concept from explicit references to external systems to the implicit references by examining where referenced knowledge is produced. Future studies with an in-depth qualitative content analysis of each reference document could further contribute to understanding the relationship between the socio-logic and the implicit reference societies.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Norges Forskningsråd [283467].

Notes on contributors

Chanwoong Baek

Chanwoong Baek is an assistant professor at the Geneva Graduate Institute. He is also UNESCO Co-Chair in Comparative Education Policy and Academic Director of NORRAG. His research interests include the politics of knowledge, comparative education policy, global education governance, and social networks.

Andreas Nordin

Andreas Nordin is professor of education at Linneaus university, Sweden. His areas of expertise include curriculum theory, education policy and comparative and international education. His recent publications include the book Taming chance in education: Control, prediction and comparison (2024), co-authored with Daniel Pettersson.

Notes

1. Only select provinces in China have participated in PISA: Shanghai (2009, 2012); Beijing, Shanghai, Jiangsu, and Guangdong (2015); and Beijing, Shanghai, Jiangsu, and Zhejiang (2018; 2022).

2. Principal Investigator: Kirsten Sivesind, University of Oslo, funded by the Norwegian Research Council [project number: 283467].

3. The complete list of all source documents can be provided upon request.

4. Because of the risk of informants to be identified, we decided against providing a summary of participants’ affiliations and occupations here to ensure their anonymity.

5. Swedish is an official language in Finland and the mother tongue of about 5% of the population.