Artificial intelligence-based learning analytics and student retention in higher education

Resumen

This research is contextualized within the growing interest in optimizing student retention in higher education through the use of artificial intelligence-based learning analytics techniques, given their potential to personalize educational strategies and improve decision-making. Its objective was to determine the relationship between such analytics and student retention, using a quantitative approach and a non-experimental cross-sectional and correlational design. The methodology combined documentary analysis of institutional data—which revealed that most students are around 25 years old, take an average of seven courses, and perform well academically, although there was evidence of heterogeneity in academic workload and economic status—with surveys aimed at measuring retention and aspects of the learning environment, work responsibilities, and personal factors. The results highlighted that more than 98% of students have medium or high levels of retention, and a high positive correlation (ρ = 0.789, p < 0.05) was identified between artificial intelligence-based analytics and retention. In conclusion, the effective use of these tools translates into the identification of risks and more informed pedagogical decision-making, strengthening student retention.