Tendencias en modelos informativos sobre la retención – deserción universitaria

The generation of data in educational environments must be used to extract relevant and timely information that improves the quality and productivity of academic institutions. This research aims to analyze current studies on student retention and / or dropout in order to determine the variables that affect them and trends in data analysis models. The guidelines of Roy Hubara and Sturm (2019) were followed to review articles from four databases, classifying them according to the approach used. The results obtained indicate that the variables that affect student retention are related to cognitive, social and organizational factors and that the tendency is to develop predictive-prescriptive models for the study of these concepts. Finally, it is proposed to develop predictive models based on statistics and learning models to improve student retention and dropout rates.