Analysis of trajectories of students in higher technical-professional education using data mining methods

In Chile, the dropout rate of students in higher technical-professional tertiary education is close to 30% in the first year. This has an enormous social impact due to the loss of funding for scholarships and free tuition, increased unemployment and increased social assistance. To understand this phenomenon, we profile students who drop out of this type of education by combining data from different sources, including their school trajectories, using different clustering algorithms. Our results identify 4 groups of dropouts, mainly distinguished by the prevalence of their gender, the fields of knowledge they study, the aid they receive, and their previous school performance. These profiles can help identify students who need support at the beginning of their studies and whose adjustment would benefit from more specific information about their career path.