Every year INVALSI (National Institute for the Evaluation of the Education and Training System) administers national learning assessments from primary to upper secondary school. In doing so, it contributes to the development of the education system in Italy. Infact, on a national scale, it provides the Ministry of Education and Merit (MIM) with information on students’ competency levels. At school level, INVALSI works closely with headteachers, teachers, and assessment coordinators to support the improvement of students’ competencies. The richness of INVALSI data lies in their large-scale coverage and detailed information at both student and school levels, enabling in-depth analyses of educational outcomes and inequalities. Among our research activities, one example concerns the development of a methodology capable of representing complex phenomena, such as school fragility, thus allowing spatial and temporal comparisons among Italian schools. A school-level composite indicator has been designed to highlight potential weaknesses within the education system and to provide a practical tool for planning interventions, due to its immediate and clear interpretability. Multiple data sources are integrated, including INVALSI and MIM data for the family context, school context and the test results dimensions. For the territorial context dimension, the Revenue Agency and ISTAT data are used. Another one involves the application of spatial statistical models for better representation of territorial disparities and to identify groups of schools in which socio-economic conditions have a greater impact on student assessments. The use of a spatial approach enabled not only the mapping of critical areas but also the identification of institutions that may be more vulnerable. This method could prove to be a relevant tool for a more efficient planning of educational interventions by adapting to the specificity of local contexts thus contributing to a deeper understanding of educational inequalities in Italy. Finally, machine learning models offer powerful predictive tools for large-scale educational data and policy interventions. A study focuses on a cohort of upper secondary school graduates enrolled in university, coming from a dataset built on the combination of different data sources, to apply different classification models for preventing dropout risk. These methods effectively identify students at risk, which could support guidance activities and policy planning. These studies demonstrate the potential of data for education not only for academic research but also for practical applications in territorial planning and policy development.

Joint Research Centre (2008). Handbook on constructing composite indicators: methodology and user guide. OECD publishing. Mazziotta, M., & Pareto, A. (2013). Methods for constructing composite indices: One for all or all for one? Rivista italiana di economia, demografia e statistica, 67, 67-80. Mazziotta M., Pareto A. (2017). Synthesis of Indicators: The Composite Indicators Approach. In Maggino F. (Ed.), Complexity in Society: From Indicators Construction to their Synthesis (pp. 161–191). Cham: Springer. Campodifiori E., Figura E., Papini M., Ricci R. (2010). Un indicatore di status socio-economico-culturale degli allievi della quinta primaria in Italia. Working paper n. 2, INVALSI 2010. Getis, A., & Ord, J. K. (1992). The analysis of spatial association by use of distance statistics. Geographical Analysis, 24(3), 189–206. Berens, J., Schneider, K., Gortz, S., Oster, S., & Burghoff, J. (2019). Early Detection of Students at Risk-Predicting Student Dropouts Using Administrative Student Data from German Universities and Machine Learning Methods. Journal of Educational Data Mining, 11(3), 1-41. Yağcı, M. (2022). Educational data mining: prediction of students' academic performance using machine learning algorithms. Smart Learning Environments, 9(1), 11.