What Are the Best Predictors of STEM Competences in PISA 2018? An Analysis of the Spanish Context Using Data Mining

Gil-Madrona P.; Guerrero-Muguerza A.M.; Infantes-Paniagua Á.; Martínez-López M. · School Science and Mathematics (2026) · DOI: 10.1111/ssm.18363

Resumen

The importance of Science, Technology, Engineering, and Mathematics (STEM) subjects for the present and future society is clear. With assessments such as the Programme for International Student Assessment (PISA), the possibility of an analysis on student-related variables predicting results in STEM areas opens. The aim was to identify the PISA 2018 variables that predict the achievement in STEM competences in Spanish students through data mining techniques. Using three data mining methods (i.e., Random Forest, support vector regression [SVR], and ridge), based on 35,943 Spanish students aged from 15.25 to 16.17 years old, 100 variables were found to predict performance in mathematical and scientific competences. Subsequent coding and classification made it possible to determine that, in mathematical competence, the variables related to access, use, teaching, interest, or skill in Information and Communication Technology (ICT) stand out. In scientific competence, these variables only represented 17%, giving way to others such as metacognition and variables related to mother tongue and reading. The personal and proximate student's context is central to performance in STEM areas.