Contributing to the question of how the global proliferation of performance-based assessment (PBA) manifests in different education policy systems and contexts, this presentation focuses on the ongoing transformation of school supervision and consultancy in German state education agencies. Germany marks an interesting case since school (policy) actors have traditionally been highly skeptical against any kind of high-stakes accountability instruments, including standardized tests or performance measures. Still, over the past two decades, also in Germany growing datafication and performance orientation have led to tremendous changes regarding instruments to monitor and support schools´ quality development (Hartong 2018). This includes the significant and ongoing expansion and centralization of data infrastructures through which various data on schools (including assessment data, but also data on student distributions, average grades, graduation rates, cancelled lessons etc.) become collected and, at least on paper, “flow into” state school supervision and consultancy, as well as a growing responsibilization of school principals for leading continuous quality development processes within their school, supervised by state authorities through steady target agreements (Bellmann et al. 2016). Building on material from a larger ongoing research project (DATAFIED, www.datafied.de), this contribution focuses on the how of transforming school supervision and consultancy in four German Länder (subnational units), examining the interactions and transactions that increasingly take place around (and through) performance data (see also Selwyn 2015), as well as the meanings that state authority actors and school principals ascribe to data tools at different contexts (such as setting target agreements). In other words, the project seeks to disentangle various aspects of decision-making and in-valuation in relation to a rising number of school performance data, including what Dudhwala and Larsen (2019) recently problematized as the “recalibration” of human judgement. The project hereby applied different qualitative methods, including a) documentary analysis, b) interviews with school supervision/consultancy actors, principals and data tool developers, as well as c) software studies of school monitoring systems. As initial findings show, also within Germany, manifestations of data-informed school supervision and consultancy are highly diverse, strongly contextualized and ambivalent, e.g. regarding what (performance) metrics data tools employ or how exactly they become linked to decision-making. These variations also include the intensity of data mediation, which on the one hand refers to the “thickness” of data infrastructures and, thus, to the flow-ability of performance data, and on the other hand to school supervisors´ (gradually changing) self-understanding of professionalism.

Author (2018) Bellmann, J., S. Schweizer, C. Thiel (2016): Nebenfolgen Neuer Steuerung unter Bedingungen von „low-stakes“ und „no-stakes“ – Qualitative und quantitative Befunde einer Untersuchung in vier Bundesländern. In: BMBF (ed.): Steuerung im Bildungssystem Implementation und Wirkung neuer Steuerungsinstrumente im Schulwesen: 208-237. Dudhwala, F., L.B. Larsen (2019): Recalibration in counting and accounting practices: Dealing with algorithmic output in public and privat. Big Data & Society, July–December 2019: 1–12.