Abstract:

The growing complexity of educational contexts and the pervasiveness of digital technologies require new ways to understand and interpret teachers' training needs, shifting instructional design toward more inclusive, collaborative, and evidence-based models. In this direction, this paper proposes a data-driven approach to identifying and analyzing teachers' professional needs, inspired by the biopsychosocial model of learning, which integrates cognitive, emotional, and relational dimensions to offer a systemic view of the educational process.

Using accessible digital tools, such as Google Forms and Excel, and automated data analysis procedures implemented in Python, a study was conducted on a sample of 452 teachers from various school levels. The questionnaire, divided into demographic-professional, experiential, and reflective sections, collected both quantitative information through closed-ended questions and Likert scales, and qualitative information through open-ended responses, allowing for data triangulation. The goal is not to provide a pedagogical interpretation of needs, but to outline an operational data management model capable of objectively and transparently supporting the decision-making process during the instructional design phase.

The developed workflow includes data cleansing, normalization, and aggregation, culminating in its visualization through graphs and summary tables. This pipeline, while based on easy-to-use tools, enables reliable and replicable analysis, laying the foundation for the application of more advanced techniques such as multivariate analysis, clustering, and the prediction of future needs. The results highlighted the presence of fragmented and poorly systematized co-design practices, difficulties in translating teaching observations into structured data, and a widespread need for digital tools that facilitate collaboration, shared reflection, and the documentation of experiences.

Needs also emerged related to supporting inclusive processes, managing relational complexity, and overcoming technological barriers, particularly in less digitalized school contexts. The data-driven approach therefore proves to be a strategic lever for strengthening the culture of data-based decision-making, fostering co-design processes that combine analytical rigor and pedagogical sensitivity.

Looking ahead, the proposed model will form the basis for the creation of a digital platform to support co-design, conceived as a collaborative environment for collecting, analyzing, and translating training needs into concrete educational actions. This evolution aims to contribute to bridging the digital divide and promoting a professional culture oriented toward sustainability, inclusion, and innovation in educational settings.

This contribution falls primarily within the scope of Research on Technology in Education and Learning Analytics, with significant implications for Teacher Training and Educational Management. It proposes a transferable methodological model that integrates data analysis and digital technologies to support the quality, inclusion, and transformation of educational processes.

Keywords:

Data-driven educational analytics, Biopsychosocial learning model, Digital co-design platforms.