To Örebro University

Over the past four decades, Sweden's education system has undergone a profound transformation, shifting from a centralised structure to a market-oriented model characterised by independent schools, deregulation, and competition. This paper introduces an innovative methodological approach to studying this transformation by applying computational text analysis with large language models (LLMs) to 45 years of parliamentary debates. By leveraging these methods and extensive parliamentary open data, we identify thematic patterns, ideological shifts, and policy discourses that have shaped the marketisation of Swedish education. Our methodological contribution lies in demonstrating how LLMs can be employed to scale up traditional discourse analysis, bridging the gap between computational methods and qualitative interpretative approaches. We engage critically with the challenges of algorithmic opacity, validation strategies, and interpretative transparency, addressing concerns about bias and the risks of black-boxed analyses. Combining machine-assisted text analysis with traditional qualitative methodologies, we present a scalable yet nuanced framework for studying education policy debates over time.

Over the past four decades, Sweden's education system has undergone a profound transformation, shifting from a centralised structure to a market-oriented model characterised by independent schools, deregulation, and competition. This paper introduces an innovative methodological approach to studying this transformation by applying computational text analysis with large language models (LLMs) to 45 years of parliamentary debates. By leveraging these methods and extensive parliamentary open data, we identify thematic patterns, ideological shifts, and policy discourses that have shaped the marketisation of Swedish education. Our methodological contribution lies in demonstrating how LLMs can be employed to scale up traditional discourse analysis, bridging the gap between computational methods and qualitative interpretative approaches. We engage critically with the challenges of algorithmic opacity, validation strategies, and interpretative transparency, addressing concerns about bias and the risks of black-boxed analyses. Combining machine-assisted text analysis with traditional qualitative methodologies, we present a scalable yet nuanced framework for studying education policy debates over time.

This article explores how the concept of school readiness acted as a boundary object in Swedish school reform between 1940 and 1975. While the era aimed for a unified "school for all," with delayed tracking, different kinds of readiness tests were widely used for early sorting. The concept's ambiguity was key to its acceptance, allowing different actors to interpret it in various ways, while still enabling coordination. This flexibility bridged scientific, political, and educational interests without requiring full consensus. Based on analyses of expert statements, laypeople's accounts, and entry testing practices, this study shows how school readiness reframes structural issues as individual deficiencies. Rather than resolving the tensions between inclusion and differentiation, the concept absorbed them, making reform implementation possible. It is here argued that the vagueness of school readiness functioned as a productive force for change.

This presentation explores educational assessment practices as techniques for governmentality and as crucial tools for the construction of the knowing subject through contemporary schooling. When the history of assessment has been written, the connection between epistemology and subjectivity has rarely been considered, but, as we will show, the epistemology in curricula contributed to the development of certain specific assessmen tpractices, which fosters particular subjectivities, through relations of power and knowledge.

We analyze the history of examinations in relation to how different assessment practices shape different abilities, such as memory, judgement, imagination, reflexivity etc, and their connection to truths and reasoning. Our discussion expands on the typical understanding of Foucault's concept of examination (1975) and introduces the organization of students as both objects and subjects of knowledge. Thus, examinations can be delineated into two categories of knowledge-producing practices: one administrative practice focused on organizing individuals, and one teaching practice through which knowledge and learning are constructed – also known as summative and formative assessments.

Our presentation builds on a genealogy of assessments in education. A close reading of Swedish curricula from the late 16th century to the early 19th century provides the backdrop for an examination of contemporary notions and practices of assessment in 21st century curricula. With shifting emphasis over time, assessments in education serve a governmental function, validating both the knowledge acquired and its epistemological foundations. This situates the subjects in an epistemological position where they are trained to develop an evaluative judgement, enabling them to distinguish basically between ‘good’ and ‘poor’ work (Boud et al., 2018). However, this subject position - a learner with control over their own learning (Nulty 2010) - becomes normalized through the everyday practice of assessments, which risks obscuring the relations between power, learning and knowledge.

We suggest that these relations are partly framed within the practices of assessment, and that the subject is tied to certain forms of knowledge, through the processes of examination, and additionally that knowledge is validated and vindicated through these practices, governing learner subjects as well as teacher subjects.

This study addresses the issue of sampling error within research on subsets of parliamentary text corpora. Two samples of parliamentary speeches relating to the marketisation of the Swedish education system, drawn thr ough different sampling techniques, are analysed and compared. The analyses find that diverging sampling methodologies can be complementary as each method adds substantial quantities of unique documents to the dataset. Further, the diverging sampling methodologies employed produce documents with similar semantic content. However, analyses of the distributi on of speeches between party affiliations and speakers indicate va st differences between the two samples. These results indicate that sampling frames can substantially influence the findings of parliamentary text analyses. We conclude that combining different sampling techniques can be a way to reduce the ris k of sampling error, which in turn can have a strong influence on the conclusions drawn from analyses of parliamentary texts. 

This paper examines the role of family members – particularly wives – within the extensive correspondence network of international and comparative education scholars documented in the unique archive of Professor Torsten Husén (1916–2009), which comprises over 80,000 pages of letters from the 1950s to the 1990s. Using a Science and Technology Studies (STS) perspective, informed by Donna Haraway’s (1988) concepts of situated knowledges and subjugated standpoints, the paper reveals how these seemingly peripheral figures actively contributed to knowledge production rather than remaining passive bystanders. By combining large language models, close reading, and preliminary network analysis of metadata (senders, recipients, greetings, references), the study shows how mentions of family construct multi-scalar relational structures. Focusing on the often-backgrounded voices at the periphery, the paper challenges traditional academic genealogies and resituates domestic spaces as co-productive sites of intellectual life. These intimate references illuminate overlooked material-semiotic entanglements, showing how wives and family members functioned as connective nodes that shaped correspondence rhythms, emotional tones, logistical support and in the end the circulation of ideas. This approach not only enriches our understanding of large-scale correspondence archives but also aligns with feminist STS calls to amplify marginal and relational perspectives, highlighting the agency of those at the edges of scholarly networks.

The field of educational science displays some specific features that are closely intertwined with temporal and spatial contexts, historical and educational knowledge, the need for reform, and scientific self-reflection. This paper investigates the impact of various contextual factors on the history of educational science in Sweden, covering the period from the 1910s to the 2020s. In the early 20th century, the first education professors primarily came from backgrounds in psychology and philosophy. Their scholarly work focused on exploring the historical evolution of educational thought, delving into influential figures such as Descartes, Rousseau, The Jesuits, and various teaching and learning practices. This early historical discourse played a key role in legitimising education as a structured domain of knowledge with a rich tradition.

Another notable aspect of historiography in education arose concurrently with the comprehensive school reforms of the mid-20th century. The story that enveloped these reforms, particularly in the 1970s and 1980s, was closely tied to the idea of ‘social engineering’ in education. This branch of historiography aimed to corroborate the reforms and the ever-growing nexus between educational science and policy. Over the subsequent decades, there was a rising interest in education within the History faculty, with a focus on micro and social aspects. However, there was little attention paid to the role and development of educational science. Currently, there is a new approach to the history of educational science that can be seen, encompassing its knowledge production from both a global, transnational perspective and a social constructivist viewpoint and thus introducing a self-reflection both in educational science and historiography.