"The data revolution comes to higher education: Identifying students at risk of dropout in Chile"

• In Event: Sustainable Urban Policy Challenges in Latin America: A 10 Year Time Horizon

Abstract

Different countries in Latin America have developed programs and public policies to increase enrollment in higher education, especially among low income and minority students. As a result, the number and diversity of students has increased, but so have the number of dropouts. Desertion is a more visible problem because the absolute number of dropouts has risen proportionately to college enrollment, and taxpayers now bear the risk of loan defaults by low-income students who drop out. In this sense, the success of educational policies focused on increasing access and equity, not only can be evaluated by the diversity of the student body in higher education, but also by the percentage of them who actually stay and finish their programs. This paper discusses how universities or central governments can use predictive models to identify students who are more likely to drop out, and thereby implement interventions to help them remain in education and obtain a college degree. We use the Chilean case to show how predictive models can target students at risk of dropping out.

Author

• Alvaro Hofflinger, Universidad de la Frontera