Policy of suspiciousness –mobilization of educational reforms in Sweden

ABSTRACT

In this article, we explore the processes of transfer and translation of education policy in a study focusing on the relationship between the Organisation for Economic Co-operation and Development (OECD) and Sweden. The purpose of the study is to investigate how selective borrowing occurs both in terms of references to different types of knowledge sources underpinning the arguments and shared discourse formations. The data were obtained from two policy reports: an OECD review report of the Swedish school system and a Swedish follow-up report proposing actual school reforms. The method was twofold: first, mapping, categorizing, and counting all the textual references in the two documents; second, critically analyzing the discourses emerging in the two policy texts. The results show that international references play a significant role in substantiating arguments for Swedish education policy. Both policy texts share a policy discourse characterized by suspiciousness toward the professionalism exercised by teachers and local education authorities.

KEYWORDS:

• Policy borrowing
• externalization
• policy discourse
• Sweden
• policy knowledge

Introduction

It has long been clear that the Organisation for Economic Co-operation and Development (OECD) is a transnational body with great influence on the formulation of education policy in numerous countries around the world. The measurements of the Programme for International Student Assessment (PISA) have continuously expanded regarding both the number of knowledge areas measured and the number of participating countries (Sellar & Lingard, 2013). There is growing research on how different countries develop education policy in relation to international standards recommended by the OECD (e.g. Lingard & Lewis, 2017; Priestley & Biesta, 2013; Takayama, 2018). There is also an emerging research field studying the kind of knowledge references that are actually used to substantiate arguments for proposed school reforms (Baek et al., 2017; Steiner-Khamsi, 2019). This article aims to contribute to this latter research field by exploring the textual references used and discourses formed to substantiate school reform in Sweden through cooperation with the OECD. There is a particular focus on policy borrowing as a selective process where nation states select what to import and how these imports are then translated within the national context (Steiner-Khamsi, 2003). In the Swedish case, this means importing discursive elements supporting the need for decontextualized policy knowledge to build a stronger and more competitive education system while delegitimizing the contextual knowledge provided by teachers and local authorities. These two lines of argumentation constitute what is discussed in this paper as a ‘policy of suspiciousness’ built up by a clearly distrustful interpretation of survey results, as opposed to a neutral or tentative one.

Purpose and research questions

The purpose of the article is to explore the bodies of knowledge constituting the argumentative foundation of ongoing Swedish compulsory school reforms initiated in close cooperation with the OECD. The study at hand investigated how the meanings of these bodies of knowledge are discursively constructed to legitimize these reforms. Although this article presents a Swedish example, a basic assumption is that the very approach driving the need for reform may look similar in many countries, while the details differ. Thus, the results of the study add new knowledge to the field of policy borrowing and lending by exploring aspects of the selection of knowledge references and education discourses that emerge when policy borrowing takes place in national politics (Steiner-Khamsi, 2019). The research questions were formulated as follows: To what extent do the references indicating the knowledge base in the examined policy documents have the character of being self-referential to the Swedish government and national authorities, and to what extent are the references external, in terms of international ones or references to research? What kinds of arguments, based on evidence from the references, are emphasized to legitimize the main proposals in the suggested reform? In this article, we analyze the references in the 2015 School Commission’s (2017) official report as well as the preceding report from the OECD (2015) to explore current patterns of references, both national and international, in legitimizing the need for a school reform The reasons for conducting this kind of analysis were, first, because of today’s political requirement for evidence-based policy (e.g. European Commission, 2017) and, second, the increasing convergence of certain ‘international standards’ provided primarily by the OECD (e.g. OECD, 2017).

The article is structured as follows. In the remainder of this introductory section, the Swedish case is contextualized. The second section consists of an account of the theoretical perspective and a methodological overview. In the third section, the results are presented in terms of reference patterns in the two analyzed documents, followed by a qualitative analysis of the dominant discourses driving Swedish educational reform. Finally, the results are discussed in a concluding reflection on possible implications.

The Swedish case

Sweden is viewed as a Western nation with a lengthy tradition of state welfare policies. However, alongside countries such as New Zealand and England, Sweden rapidly moved toward more decentralized and market-oriented arrangements for education during the 1990s (Lubienski, 2019). Policy borrowing in the Swedish context has been described as a historically silent process of importing undeclared ideas and concepts (Ringarp & Waldow, 2016). It is probably most appropriate to talk about silent borrowing from 1994 onward, when Sweden became a member state of the European Union (EU), and ‘silent’ is used to refer to silence regarding EU policy (Nordin, 2012). However, for the OECD, rules other than those of the EU apply. The OECD is not viewed as a political project in the same way that the EU is but, rather, as a source of neutral experts available to the Swedish government, as evinced by how the government presented its mission to the OECD to Swedish media:

After the PISA report condemned the Swedish school results, the government is now taking the initiative of an independent international school commission which is reviewing the Swedish school system … ‘It will consist of the best school experts in the world to be appointed by the OECD, which conducts the PISA survey’, said the Minister of Education and Research Jan Björklund (the Liberals) at a press conference this morning. (News on Swedish television, 14 January 2014)

A consequence is that it is not only uncontroversial but also legitimizing to borrow international policies from the OECD by making explicit references to the PISA results and to OECD expert reports (Wahlström, 2018). The 2012 PISA results indicated a performance downturn that reinforced strong domestic political criticism of the Swedish school system. The critique had been going on throughout the 2000s due to decreasing knowledge achievements in international knowledge comparisons, especially in the PISA surveys. By bringing together statistical evidence, ‘often in the form of emotive graphs of a spiral decline’ and by creating a discourse of the need for change, ‘the OECD acted as the negotiating arena for both those believing in data for improvement, as well as those actors identifying the need for reform as more of a political and ideological matter, rather than a scientific one’ (Grek, 2020, p. 2). In a crisis, due to the school’s inability to offer students relevant knowledge, the OECD becomes an expert that can provide Sweden with authoritative advice on possible ways out of the crisis.

Thus, in 2014, the Swedish government commissioned the OECD to review the quality of the Swedish national compulsory school system. The purpose of the evaluation was to identify reasons for the decline in the Swedish students’ knowledge achievements, draw on lessons from other OECD countries based on the PISA results, and suggest policy areas for reforms (OECD, 2015). At the same time, the Swedish government announced that it would appoint a school commission to put forward a concrete proposal, based on the OECD’s recommendations, for reforms for the Swedish school system. Thus, after receiving the OECD report, the Swedish government assigned the 2015 School Commission the task of proposing school reforms with the aim of increasing knowledge results as well as teaching quality and equivalence between schools.

Theoretical framework

Over the last decade, references to the OECD and PISA results have become an increasingly important strategy for legitimizing political reforms in Sweden (Nordin, 2017; Ringarp & Waldow, 2016; Wahlström, 2018). To capture these cross-national processes of policies traveling between different contexts, the concepts of ‘policy borrowing’ and ‘policy lending’ have been established as theoretical terms within the broader field of comparative policy studies. Today, the focus in this research field has shifted from bilateral to international frames of reference because of the increasing prominence given to ‘international standards’ and ‘best practices’ (Steiner-Khamsi, 2012). Drawing on Steiner-Khamsi’s (2012) research on policy borrowing, this study explored the process of policy borrowing for international reform suggestions and bodies of policy knowledge built on best practices and international standards from a local policy context, represented by the educational policy processes in Sweden during the years 2015–2017. The elements of policy borrowing that were of specific interest in this study were transfer and translation (Cowen, 2009). The aspect of ‘transfer’ refers to how educational ideas are formed and communicated in transnational policy spaces. The aspect of ‘translation’ represents the re-interpretation of transnational educational ideas in the context of an individual country. In this article, the term ‘transnational’ refers to international organizations with national governments as members, such as the OECD and the EU.

According to Grek (2013), international testing of students’ knowledge has offered policymakers with data to govern while at the same time almost fusing the realms of knowledge and policy. This fusion of data brings together expertise with advocacy for undisputed, universal policy solutions into a single phenomenon that can be expressed as international standards. Therefore, the term ‘international standards’ is actually an expression for a mixture of different policy lessons learned from diverse national cases of best practices and international evaluations, which have merged together into common standards. References to international standards are, thus, perceived as decontextualized and denationalized measurements of knowledge. The process of decontextualization takes place through the dissemination and dilution of what were originally specific policy cases to a large number of countries. To strengthen the argumentative basis for reforms, domestic policy actors tend to build discourse coalitions together with external actors to bridge disparate national interests and convince policy actors of the need for certain reforms (Steiner-Khamsi, 2016). Discourse is understood here as exchanges of ideas whereby people deliberate and reflect on policies to coordinate the direction of policy reforms. To develop or change education policies, people build discourse coalitions for reform against established interests by coordinating their communication and knowledge sources (Schmidt, 2011).

The translation of international policy bodies of knowledge within an individual country is not a straightforward, top-down process from a transnational level to a national one, since policy borrowing is never a matter of ideas as a whole. Instead, policy borrowing is always selective in nature (Wahlström, 2020). The processes of translation between different policy arenas are both about learning from and benchmarking against actual international bodies of knowledge and about shaping multiple discourses of policy knowledge for the national legitimization of certain reform proposals. The term ‘policy knowledge’ represents complex processes of well-probed beliefs, information, and the utilization of knowledge claims in dealing with policy problems. Knowledge is interpreted in this article in a broad sense of information rather than as rigorous evidence-based knowledge. Policy knowledge underpins the arguments of suggested policy solutions, where ‘[t]he acts of producing and using information in organizational decision-making have the symbolic value of expressing the perceived rational foundations of choices’ (Radaelli, 1995, p. 162).

Drawing on the work of German sociologist Niklas Luhmann, Schriewer (2003) emphasizes the importance of the concept of ‘externalization’. Externalization involves ‘the selective description and synthesising interpretation of international phenomena for issues of educational policy or ideological legitimization’ (Schriewer, 2003, p. 276). This term should not be understood as a description of international policy evidence or historical policy developments as such but, rather, as filtering the reception and interpretation of international policy movements in relation to the internal context of a given system. Thus, externalization is a consequence of the need for domestic education systems to reach out to other systems in their self-referential and self-reflective processes of knowledge building. National education policy systems thereby create supplementary meanings of international policy knowledge in accordance with their own cultural, political, and ideological settings (Schriewer & Martinez, 2004). This logic of intrasocietal reflections of bodies of knowledge emphasizes a multiplicity of meaning processes in multiple worlds rather than in one world. Therefore, the concept of externalization ‘calls for particular attention to the interpretative reception and transformation within the educational discourses of different nations’ (Schriewer, 2003, p. 277)—that is, in the phase that Cowen (2009) refers to as translation. Externalization to international systems of policy knowledge allows for a high degree of selecting which knowledge to receive and how to semantically understand and use it, contributing to the shaping of idiosyncratic discursive spaces (Schriewer, 2003). Steiner-Khamsi (2014) argues that externalization, in terms of references to international standards or international knowledge, occurs in situations of domestic policy contestation for the legitimation of certain reform proposals.

The phenomenon of policy borrowing is a conceptualization of a long history of international interactions between different sectors in society. The concept of policy borrowing in education policy, based on a theory of externalization, includes power relations revealed by which system is perceived as representing a desirable ‘reference society’ from which to borrow ideas, policies, and practices. Drawing on Steiner-Khamsi (2014), the concept of policy borrowing includes the following characteristics: (i) terms such as ‘international standards’ and ‘best practices’ resonate with particular moments of national agenda setting for education policy; (ii) policy borrowing has the function of a national coalition builder because it allows local parties with conflicting opinions to agree on the borrowing from an external, supposedly more neutral, policy actor; and (iii) policy borrowing is selective in nature because the reception of policies is influenced by context-specific reasons within which certain elements in a policy from elsewhere become emphasized in the transfer and translation from an international context to a national one. As Steiner-Khamsi (2003) points out, borrowing selectively from elsewhere can sometimes become part of a national discourse of distrust, or with Steiner-Khamsi's (2003, p. 2) term ‘scandalization’ by ‘highlighting the weaknesses of one's own educational system as a result of comparison’. Thus, borrowing from elsewhere can be used as a strategy for legitimization as well as delegitimization of a national policy agenda. In this study, suspiciousness was understood as part of such a strategy of distrust. Acting suspiciously against one’s own national education system becomes part of a political strategy to allow national politicians to maintain public legitimacy. Directing suspicion against one’s own national education system, while simultaneously borrowing policy (and thus legitimacy) from elsewhere can be seen as an expression of the destabilization of a nation state as the natural site and scale of politics in a globalized context (Lingard & Ozga, 2007). It is reasonable to believe that as the influence of the international policy arena increased during the 2000s, externalization increasingly became an integrated part of Sweden’s national policy.

Here, Sweden has the role of a borrower of education policies from the OECD, which is the main lender. Sweden’s role as a borrower of OECD education policy is voluntary in the sense that Sweden chose to turn to the OECD to legitimize its forthcoming school reform.

Methodological considerations

The present study had a mixed-methods approach combining elements of quantitative and qualitative research. The data analysis included a quantitative approach in an initial analytic phase involving mapping, categorizing, and counting bibliographic references in the reference lists of the two analyzed documents. The second sequence in the analysis built on a qualitative research methodology—namely, critical discourse analysis. This combination of methodological approaches can be termed a sequential quantitative–qualitative design, with the qualitative approach as the priority (Cresswell, 2010).

The data were derived from two policy documents: the OECD (2015) report and the School Commission (2017) report. Both reports were carried out on behalf of the Swedish Ministry of Education and Research. The task for the OECD was as follows:

• (1) identify the main reasons for the decreasing trends in Swedish students’ performance; (2) draw on lessons from PISA and other benchmarking countries/regions with an expert analysis of key aspects of education policy in Sweden; and (3) highlight areas of policy and its implementation which might add further value to Sweden’s efforts to improve student performance. (OECD, 2015, p. 14)

The directives from the Swedish government to the School Commission outline that the latter has the task of suggesting proposals ‘aimed at improving knowledge outcomes, improving quality in teaching and increasing school equivalence’ (School Commission, 2017, p. 385). Moreover, the proposal should ‘be based on the OECD final recommendations from the thematic review of the Swedish school system’ (p. 385). Given the importance the government attributes to the OECD (2015) report, we characterize both it and the 2015 School Commission (2017) report as the two main sources underpinning the policy guiding Sweden’s ongoing school reform. Together, the two documents comprise 652 pages.

Both reports contain bibliographies—that is, references to different knowledge and information sources utilized by the authors of the policy documents. First, all references in both documents were documented and categorized into three groups: domestic, international, and Nordic (see Baek et al., 2017). Second, all references were coded by the type of document: governmental, academic, report, etc. Third, the number of times a reference occurred in the document was counted to get an idea of the importance of a certain knowledge source for the specific policy document. Lastly, the textual references cited in both policy documents were identified. However, this mapping of the total number of references used as knowledge sources only shows part of the picture of the transfer process between an international and national policy arena. In this particular case, the Swedish government explicitly stated that the School Commission’s proposal should be based on the OECD report. It is, therefore, reasonable to believe that policy discourses from the OECD (2015) report permeate the School Commission’s (2017) report to a higher extent than can be shown by merely examining references indicating the utilization of certain bodies of knowledge.

Therefore, in the second sequence of the analysis, the policy discourses in the two documents were analyzed to capture the broader basis for the argumentation of policy proposals from the 2015 School Commission. For the analysis of underpinning policy discourses, a critical discourse analysis was conducted. Critical discourse analysis involves analyzing dialectical relations between discourse and socio-cultural practices related to the discourse as well as the internal relations of discourse. This analysis accounted for the fact that that the social world can be construed in different ways, which implies that discourses always include power relations (Fairclough, 2010). First, both the OECD report and the School Commission report were read in their entirety. Thereafter, the analysis centered on the relationships between socio-cultural practices and education policy proposals to capture both overlapping and diverging policy discourses in the two reports. Examples of socio-cultural practices present in the two policy reports are governance of the national school system, the PISA surveys, the teacher education system, the teachers’ assessment practice, and the teachers’ status in society. Thus, in a third reading of the two reports, the discursive understanding and interpretation of policy problems and policy solutions in relation to these kinds of social practices were in focus. With reference to Fairclough (2010), language itself should be analyzed as a social practice. The analysis in this third step, thus, centered on how the language in the reports forms power relations and argues for what is presented as common sense, as well as what count as reliable knowledge sources. The analysis of intertextual and interdiscursive elements in the two policy documents revealed the micro processes of discursive transfer and translation (Cowen, 2009) between international and national policy arenas. Moreover, the production of the two documents as such represents what Fairclough (2010) terms a discursive practice, which includes observing the weight given to the sender of the texts, references to other texts, issues addressed, and expected audience.

Explicit but selected borrowing

The report from the 2015 School Commission (2017) includes 337 references in total. The identification of textual references showed that the majority, 74%, are domestic and less than 3% are regional (Nordic), while 23% of the references are international (Table 1).¹

Table 1. The geographical origin of references in the Swedish Official Report (2017) (n = 337).

Region	Frequency	Percent
Domestic (Sweden)	250	74.2
Regional (Nordic)	9	2.7
International	78	23.1
Total	337	100

In a corresponding mapping consisting of data from the OECD (2015) report, the majority of the references, 78%, are international, 20% are Swedish, and 2% are regional (Nordic). In total, the OECD report contains 223 references (Table 2).

Table 2. The geographical origin of references in the OECD (2015) report (n = 223).

Region	Frequency	Percent
Domestic (Sweden)	44	19.7
Regional (Nordic)	4	1.8
International	175	78.5
Total	223	100

An examination of the type of document referenced in the School Commission (2017) report demonstrated that the dominant pattern of references is domestic policy documents from national authorities (n = 117) and government documents (n = 90), comprising mainly official reports. The references to reports from transnational organizations, mainly the OECD, are fewer (n = 15).

A corresponding analysis including data from the OECD (2015) report showed that there are primarily international references to reports from transnational organizations (n = 81), of which 64 references refer to OECD publications. Among the references in the OECD report, 30 refer to reports from national authorities and 34 refer to international peer-reviewed studies.

In the OECD report, references to its own organization, as well as to other transnational organizations, dominate. In the School Commission report, most of the references represent sources from national authorities and the Swedish government. Of a total of 560 references in the two policy documents, only seven references are included in both reports. The mapping of references provides insight into the logic of the knowledge base in the policy arena. The sources of knowledge might, however, be more complex than they appear at first glance. Thus, it is also interesting to look at the frequency of each reference.

The weight of certain citations

In its report, the 2015 School Commission (2017) refers 58 times to reports from the Swedish National Agency for Education (NAE). There is a pattern of one to five references to each individual NAE report; the most common number is one to two references to each report. However, there is one exception to this pattern. The School Commission refers 12 times to one of the reports from the NAE. The report is a version in Swedish of the OECD (2014) report, TALIS 2013 Results: An International Perspective on Teaching and Learning.² Thus, even references that appear to be domestic can still have content produced by transnational organizations. The most cited reference in the 2015 School Commission report is the OECD (2015) report, which is referred to 44 times.

The most cited sources in the OECD (2015) report are 53 references to four OECD reports related to the 2012 PISA results. The 2012 PISA survey was the most recent PISA test at the time. It was also the PISA test on which Swedish 15-year-olds scored their lowest results among all PISA tests so far. Alongside the references to PISA documents, two other documents are prominent in the OECD (2015) report. One is the TALIS 2013 report (OECD, 2014). This document is referred to 28 times. In the TALIS report, the importance of teachers in raising students’ knowledge scores is emphasized. The other prominent report is OECD Reviews of Evaluation and Assessment in Education: Sweden 2011 (Nusche, Halász, Looney, Santiago, & Shewbridge, 2011). This report is referred to 24 times. Together, the most cited reports in the OECD (2015) report—the PISA results, the TALIS report, and the OECD country report on Sweden—shape an understanding of the Swedish school system deeply rooted in an OECD discourse on reasons for school reform.

To gain deeper insight into the meaning of the policy behind the reform, the examination needed to include an analysis of the dominant discourses driving the reform. The following section presents the results of a critical discourse analysis showing the four main discourses that are common to both reports.

Translation of international standards in the national arena

In this section, four discourses on the relationships to socio-cultural educational practices building up pressure for Swedish school reform are presented: teachers’ professional practice, governance of the national school system, the role of the school in society, and assessment practices.

The discourse of distrust

In the NAE (2014) report on the TALIS survey, a dominant theme is that only 5% of Swedish teachers feel their profession is appreciated by society. Moreover, half of the teachers would not choose to become teachers again if they got a second chance. The NAE states that Sweden is one of the countries where teachers are least satisfied with their professional choice, both compared to other Nordic countries and the average for EU/OECD countries. At the same time, the NAE notes that most Swedish teachers provide a very positive picture of their wellbeing at the schools where they currently work and of their own efforts as teachers. They also experience to a greater degree than other TALIS teachers that they can make their students feel they can do well in their school work. However, the NAE does not appear to trust the information about the wellbeing and sense of success expressed by the Swedish teachers, as indicated in the following question: ‘Might this stem from some kind of low expectations on the students?’ (NAE, 2014, p. 90).

A similar example comes from the underlying OECD (2015) report, where some results from the TALIS 2013 survey are presented. According to the OECD, the majority of the Swedish teachers (96%) report that they are satisfied with their performance in their current school, which indicates a high sense of self-efficacy. The report states that ‘[i]t is difficult to reconcile the high sense of self-efficacy of teachers with the low performance of Swedish students in the various international assessments’ (OECD, 2015, p. 112). The OECD also shows greater confidence in the PISA results than in the teachers’ self-perception, noting that teachers, according to the PISA survey, consistently seem to have problems dealing with disruptive and disengaged students (p. 112). How, then, can the teachers be pleased, the OECD implicitly asks. The thought that PISA measurements may not be completely reliable in terms of all their different elements does not seem to have struck the authors of the OECD report.

There are clear elements of suspicion that shape one of the main policy discourses in the 2015 School Commission (2017) report, which we term ‘the policy of distrust’. Even when the results show some positive tendencies, as in the example of teachers’ experience of satisfaction with their work in the classroom cited above, the OECD evaluators show greater confidence in the negative results from the PISA survey. Negative results are not questioned in the report, while potentially positive results from the PISA survey or other sources provoke suspicion. In both the report from the School Commission (2017) and the preceding OECD (2015) report, the survey results from international comparisons in the PISA and TALIS are used as unquestioned yardsticks, even though parts of the surveys deal with value issues and estimates. The policy discourse of distrust contributes to maintaining a broader background discourse of crisis and a school persistently in need of profound changes—a discourse nourished by large-scale assessments with the dominant ideas of ranking and competition as the way to national success.

The discourse of standardization

In its report, the School Commission (2017) refers 29 times to the OECD (2015) report. A main theme in the OECD (2015) report is the problem of declining knowledge, according to the 2012 PISA survey, in combination with disbelief regarding a decentralized school system with the municipalities as those primarily responsible for the schools. The OECD (2015) lists the shortcomings, comprising decreasing knowledge performance ‘in all PISA domains’ (p. 7), a high level of truancy, a lack of perseverance in students’ learning, and teaching as a low-status profession. The OECD concludes that the current ‘highly decentralised and complex system’ (p. 56) is unclear and unable to deliver the desired results. According to the OECD, the priorities are indistinct and the implementation ineffective. The report does not relate the municipal responsibility to the matter of local democracy but instead refers to the OECD’s own sources of international benchmarking, arguing for the need to build an effective national strategy for governing the school system in Sweden so that national strategies have a distinct impact at all levels of the system, with clear assessment and evaluation arrangements.

The policy discourse of standardization is characterized by a push for uniformity and standardization as an expression of the efficiency, equivalence, and clear governance of the school system at the expense of local influence and local variations. Echoing the OECD’s view of the Swedish school system’s shortcomings and the need for a more efficient governance system for the schools, the School Commission (2017) report proposes increased state governance, with the reinstatement of a state regional authority and targeted state subsidies. The Swedish response to the transnational policy of standardization, thus, becomes a question of demands for increased state governance.

Referring to ‘international standards’ is not the same as benchmarking against some absolute measurements. Policy as standardization should rather be understood in terms of a policy discourse shaped and maintained by international policy organizations, primarily the OECD. The knowledge base for international standards is developed by international knowledge measurements, such as PISA surveys, and international evaluations more generally, such as the TALIS project. Although numbers and measurements legitimize the discourse of standardization, it is still a vague discourse. The OECD communicates the policy discourse of standardization and uniformity to its member states through country reports and other international evaluations. In response, the member states relate their national policies to the international discourse while adapting different elements of the standardization discourse to their own national contexts in the translation process.

The discourse of low expectations

The School Commission (2017) refers both to the OECD (2015) report and to the TALIS 2013 Results (OECD, 2014) when discussing the matter of an unsatisfying study environment. A common problem surfaces in the annual attitude measurements from the NAE and the Swedish School Inspectorate, according to the School Commission (2017). The School Commission suggests that the principal’s responsibility for safety and a satisfying study environment should be clarified in the national curricula. The curricula should also state that students should show respect and consideration for teachers and for other students, and it should emphasize the importance of students taking responsibility for their learning through their own efforts and for contributing to a supportive school environment.

The OECD (2015) report describes a generous Swedish welfare state, increasingly accepting of benefit dependency and of the drifting of social norms and values as a background to the OECD’s general feeling of ‘complacency throughout large parts of the school system’ (p. 19). The OECD argues that this national complacency has negative effects on how teachers manage to ‘deliver the curriculum effectively’ (p. 71). Further, the OECD argues that students are not being challenged enough and are, therefore, not reaching their full potential. To create an effective school system, measures against disciplinary problems are argued for as a fundamental prerequisite. According to the OECD, the lack of discipline is not only a problem related to the attitudes of students and their parents. The OECD argues that the identified disciplinary problems are due to the teachers’ too low expectations. When analyzing the state of the teaching profession, the OECD report (2015) relies heavily on the TALIS 2013 report (OECD, 2014).

In this discourse of low expectations, Swedish teachers are described as working alone, without a mentor, and with few possibilities to get feedback on their teaching. Altogether, the OECD (2015) report constructs a discursive theme on a failure to uphold discipline and high ambitions among students, parents, and teachers—albeit without giving any explanations or reasons for why this would be the case, except a hypothesis of ‘complacency’.

The discourse of comparability

The OECD (2015) report cites the Swedish guidelines: ‘when setting grades, teachers take into account all accessible information about the student’s proficiency in relation to national knowledge requirements, and make an all-round assessment of the proficiency the student shows’ (p. 24). However, the OECD presents this approach of context-dependent information among the many sources of information as a problem because it complicates comparability. Due to the hegemonic position of comparability in the OECD context, contextual information is assumed to distort the ideal of objectivity and impartiality underpinning large-scale assessments, such as PISA and TALIS, and therefore is thought of as something that should be avoided. There is no discussion of the fact that assessments based on multiple sources instead could be seen as a sign of professional strength or something that actually offers a more valid basis for the individual student’s knowledge progression.

The 2015 School Commission (2017) follows up the theme of comparability from the OECD (2015) report by noting a failing capacity and lack of responsibility for evaluation among many local school authorities and those responsible for independent schools. The School Commission argues that the state has underestimated the demands placed on a results-driven school system with regard to the capabilities of those responsible for following up, evaluating, and analyzing their results. Nor has the state ensured access to necessary information for the entire school system to make it possible to require accountability at all levels of the system. The NAE and the Swedish Schools Inspectorate have repeatedly stated that many local school authorities and those responsible for independent schools lack the capacity and the necessary documentation to be able to analyze and remedy deficiencies and problems in school activities (School Commission, 2017).

In the discourse of comparability, the OECD advocates a clearer division of who is to be held accountable for what across the school system and traces the lack of accountability to the early 1990s when the 290 municipalities took over the responsibility for schools from the state. The OECD (2015) report identifies the variation in policy enactment among the municipalities as a problem that has to be dealt with. As in the case of teachers, it is once again the contextual premises influencing the implementation of reforms that are identified as the problem. According to the two policy reports, the decentralized school system launched in the early 1990s, with the aim of strengthening both the effectiveness of the system and the local democratic potential, has resulted in too many differences between municipalities. Expressed otherwise, the ideal of education is perceived as contextually independent regarding the way education is planned, realized, and assessed. Thus, measurability becomes a self-reinforcing and highly normative discursive theme within the policy of comparability. The measurements prescribe what is perceived as desirable education, regardless of contextual conditions or judgments by the professional actors involved.

A discourse of suspiciousness to substantiate the need for reform

The aim of this study was to explore in detail what references constitute the knowledge base and argumentative foundation of ongoing Swedish educational reforms and to investigate how the meanings of these references are discursively constructed to legitimize the educational reforms (Steiner-Khamsi, 2012, 2016). Mapping and categorizing all the references in the two reports clarified that the OECD (2015) report mainly refers to other OECD publications, while the School Commission (2017) report primarily refers to documents produced by the Swedish government and national education authorities. Thus, the first conclusion is that the two main policy documents in this study are highly self-referential. This means that the knowledge sources are largely internal, related to the organization itself or to organizations close to the publishing organization. Nevertheless, it seems clear that the two documents are closely related to each other. How, then, does the transfer between an international and a national policy actor work?

To gain deeper insight into the impact of different knowledge sources, the frequency of the references also needed to be measured. Both in the OECD (2015) report and the School Commission (2017) report, references to the OECD dominate. The second conclusion is, therefore, that references to the OECD play a significant role in substantiating arguments for Swedish education policy. Supported by the Swedish government, the OECD has established itself as a neutral expert body in relation to Swedish public opinion. By conducting international evaluations, primarily PISA and TALIS, the OECD has achieved an authoritative position as experts on what is characteristic of a successful school system. Such a position gives the OECD the opportunity to assess individual school systems while at the same time linking its proposed solutions to a preferred policy approach (Grek, 2020). This shows that externalization (Schriewer, 2003; Steiner-Khamsi, 2014) can be very effective when national ideological policy interests strive to exert reform pressure for a certain policy solution. It simply becomes difficult for domestic ideological opponents, and also for researchers, to go against the ‘OECD machinery’ by highlighting other results or sources of knowledge than, for example, those of the PISA survey.

However, it is important to remember the character of mutual exchange in the transfer of policy ideas, including shared interests and selective policy borrowing. Everything is not orchestrated from the level of the OECD and everything is not borrowed (Wahlström, 2018, 2020). In this study, the focus was on the Swedish government’s utilization of the OECD’s evaluation and policy proposals to strengthen its own policy in relation to domestic opposition claiming other sources of knowledge within policy and research. The results show how the translation of certain discourse elements are interactively reinforced by a shared emphasis of specific policy elements claiming that there is a crisis within the Swedish school system that needs to be reformed (Nordin, 2019). The moral assumption underpinning the four policy discourses identified in this study is that you cannot perform at the highest level if you are already satisfied with your results (complacency vs. competitiveness). The rational assumption supporting the reform pressure is that ‘pure’ data constitute the only reliable basis for assessment and evaluation (professional judgments vs. large-scale evaluations). What the four discourses have in common, built on both moral and rational assumptions, is that they, taken together, express doubt about the professionalism of teachers and the decentralized national school organization. The reasons for doubts appear in cases where teachers’ views on the national school system do not match the opinion of the OECD. In these cases, the OECD and the School Commission doubt the appropriateness and efficiency of the prevailing discourses. Both the transnational and national policy arenas shape their arguments for school reforms through a policy of suspiciousness directed toward teaching professionals and local education authorities, which suggests that it is reasonable to expect school reforms to focus on the areas of students’ knowledge achievements and increased state control.

Acknowledgement

We thank the research project ‘Policy Knowledge and Lesson Drawing in Nordic School Reform in an Era of International Comparison,’ funded by the Norwegian Research Council and led by Kirsten Sivesind, Oslo University (project number: 283467), for insights and support constituting the basis for this article. We also thank Marie Hallbäck, Linnaeus University, for her work on encoding the documents.

Disclosure statement

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

Notes

1 The coding system was developed within the project, Policy Knowledge and Lesson Drawing in Nordic School Reform in an Era of International Comparison, funded by the Norwegian Research Council. Project Number: 283467.

2 Teaching and Learning International Survey (TALIS).