Abstract

Teacher well-being is closely linked to the concept of well-being developed in social psychology. This is a polysemic construct that involves a complex interaction of psychological processes, such as positive and negative affect, personal engagement, and sustained life satisfaction. In this study, teacher well-being is addressed through the application of two complementary scales: the WEMWBS, which assesses overall well-being, and the PANAS, which measures both positive and negative affect. Both instruments were administered to a sample of 60 teachers. AsAs a methodological contribution, we propose a strategy based on unsupervised learning to identify priority groups. This approach includes the development of a web application prototype aimed at data analysis and interactive exploration of the results. Based on The average scores from the WEMWBS and the negative subscale of PANAS, a hierarchical cluster analysis was conducted, which allowed the identification and characterization of three groups: (1) Low well-being with contained affect, (2) Stable or moderate well-being, and (3) Moderate well-being with emotional overload, The latter being the highest-risk group.