Educational Data Science and Machine Learning: A Case Study on University Student Dropout in Mexico

Nowadays, university dropout is a disturbing phenomenon that affects students, educational institutions, and the state. A look at this phenomenon from Educational Data Science and the application of Machine Learning techniques allows us to search for the potential permanence of the students, which is why this research aims to predict school dropout in the first year of studies. university level using these techniques. A practical case study is analyzed in the educational field using a private university student database in Mexico. It is shown in the study that the metrics and the visualization of the structuring to analyze patterns allow to determine that the characteristics that best predict institutional dropout in the first year of studies at the university level are the average of the student in the first period and the percentage of the scholarship.