Education policies that work: A contextsensitive ‘big data’ approach: EDUPOL

Programme(s): HORIZON.1.1 - European Research Council (ERC)

Topic(s): ERC-2022-COG - ERC CONSOLIDATOR GRANTS

Call for proposal: ERC-2022-COG

Funding Scheme: ERC - Support for frontier research (ERC)

Grant agreement ID: 101086663

Objective:

In the last decades, profound socio-economic changes have posed unprecedented challenges to the traditional centralised model of state-provided education. This has led many countries to adopt neoliberal educational policies (NLPs) promoting school autonomy, competition and accountability as leverages for school effectiveness. Yet, given increasing societal inequalities, we are also witnessing a resurgence of democratic inclusive educational policies (DIPs) with egalitarian aims. Although these different policies can coexist, they have so far been scrutinised separately. EDUPOL develops the first comprehensive comparative analysis of the interplay between NLPs and DIPs. By building on the concepts of policy configurations and context-specific effects, the project proposes a paradigmatic shift from the focus on ‘what works’ to ‘what combinations of policies work for whom and in which circumstances’. It pursues four key aims: 1) to investigate the trends and diffusion of NLPs and DIPs across the globe in the last four decades; 2) to assess their effects on quality and equality in students’ (cognitive and socio-emotional) competencies and civic engagement; 3) to map within-country school-specific policies; 4) to uncover the heterogeneous effects of school-level policies across schools and territories. EDUPOL builds a novel macro-longitudinal dataset of educational policies and integrates insights from international repeated cross-sectional data, groundbreaking Italian big data (millions of student records, textual data from school reports, administrative archives, satellite data), student population panel data and qualitative interviews with key school actors to achieve its ambitious objectives. By combining qualitative comparative analysis, spatial panel regression, flexible machine-learning techniques and causal-inference methods, the project will significantly advance our understanding of the conditions under which educational policies work.