Student Support System Higher Education Predictive Models

Currently, CEO ETHNODATA, PhD candidate in Complex Systems Engineering, Mg. in Statistics, Mg. in Territorial Environmental Management and Forestry Engineer. My work and academic development is related to teaching and research in mathematics and statistics for more than ten years at the Technological University of Chile (INACAP).

The main proposal of this research is the development of predictive models, based on input variables (sociodemographic) and academic variables for the identification of students at risk of failing mathematics subjects. In INACAP, mathematics subjects have a failure rate of close to 40% in the last 4 years, and as a consequence of this figure, student desertion increases. This study considers three mathematics I subjects (MTAN02, MTES01 and MTIN01) that are equivalent in terms of their contents, are present in various careers and have national coverage. One of the important issues to be addressed is the development of an "early warning" system for those students at risk of failing, with the aim of providing tools and support to reduce this risk in a timely manner, the academic variable considered key is "grade 1", this variable in conjunction with student characterization variables allows the estimation of the risk of failing and at the same time manage the student's academic support.