Aprendizaje Basado en Datos: estrategias predictivas para mejorar la intervención pedagógica y la toma de decisiones en el contexto educativo ecuatoriano

Digital transformation has generated new opportunities to optimize educational processes through Data-Based Learning (DBL). In this context, this study analyzes DBL predictive strategies to improve pedagogical intervention and decision-making in the Ecuadorian educational system. For this purpose, a systematic review of 15 studies published between 2022-2024 was conducted, applying PRISMA criteria and methodological quality assessment. The results reveal that predictive models achieve 85-92% accuracy in early identification of dropout risks, while personalization systems improve academic performance by 23% and student retention by 18%. Furthermore, four critical dimensions for successful implementation were identified: technological, pedagogical, organizational, and evaluative. Nevertheless, significant challenges persist such as technological infrastructure limitations (73% of studies), insufficient teacher training (58%), and ethical concerns about data privacy (41%). In conclusion, DBL represents a transformative tool for Ecuador, however, it requires strategic investment in infrastructure, teacher competency development, and specific regulatory frameworks. Therefore, a gradual implementation model is proposed that leverages national context opportunities to achieve sustainable and effective adoption.