Objectives Despite having a comprehensive model of education, Norway has experienced a high number of students dropping out of secondary education. 20% do not complete within the five years that they have a legal right to access upper secondary education (Statistics Norway, 2024). Even if the completion rate has increased recently, this represents a significant societal and political problem. This paper analyzes which models and strategies of district administrators can be identified as particularly effective as dropout prevention routines and how these routines are transferred and translated across contexts. Theoretical perspectives The analysis draws on institutional work to understand the connection between actors and established structures and routines across levels and institutions (Lawrence et al., 2006). A particular focus is directed towards ‘successful’ knowledge sharing (Røvik, 2023; 2016), which in our context is about the conditions for translating and transferring well-functioning routines across administrative levels and institutions and involves two critical phases: decontextualisation and contextualisation. The first phase involves translating a desired practice in a specific organizational context into an abstract representation (e.g., texts) while ensuring the relevant information needed to explain and understand how the practice works in a specific context. The second phase, contextualisation, concerns the complex translation that takes place from an abstract representation of a desired practice to a recreated new practice in another organization. Methods and data sources For this qualitative analysis, we have worked both inductively and deductively (Ragin, 2023) by first identifying key routines that inhibit dropout. They appeared as the most prominent observed routines in terms of influencing completion rates and were also emphasized by the informants. We first identified examples of decontextualisation and contextualisation that can be characterized as ‘successful.’ By selecting ‘best cases’ for further analyses, we had the opportunity to develop a greater understanding of the routines and knowledge-sharing processes. Key documents (e.g., strategy plans, reports), interviews with 20 administrators and staff from three districts selected according to geographical location and achieved completion rates over time, and 55 hours of observations of relevant meetings were analyzed. The data were gathered over three school years (2021-24). Results Three knowledge-sharing routines have been identified: a) new, procedure-driven routines and tools that mask variation in routines, b) use of dropout and grade data for accountability, and c) a moral imperative of ‘the pupil's best interests,’ fostering interdependence among actors.

Blumer, H. (1954). What is Wrong with Social Theory? American Sociological Review, 19(1), 3–10. Lawrence, T. B. & Suddaby, R. (2006). Institutions and institutional work. In: The Sage Handbook of Organization studies, Vol. 2. (s. 215-254). SAGE. Ragin, C. C. (2023). Analytic Induction for Social Research. University of California Press. Røvik, K. A. (2016). Knowledge Transfer as Translation: Review and Elements of an Instrumental Theory. International Journal of Management Reviews, 18, s. 290-310. Røvik, K. A. (2023). A Translation Theory of Knowledge Transfer: Learning Across Organizational Borders. Oxford Academic. Statistisk sentralbyrå (2024). Gjennomføring i videregående opplæring – SSB