Academic risk model through the analytical hierarchical process (AHP)

Ana Oña ^1* Carmen Gloria Burgos ² Jorge Castillo ² Manuel Molina ² Milena Murillo ²

• ¹ Swiss Paraplegic Research, Nottwil, Switzerland
• ² Universidad de Atacama, Copiapó, Chile

Higher education institutions, especially those reliant on public funding, focus heavily on academic performance and student dropout rates. Early identification of students with academic challenges is crucial for timely intervention. However, many predictive models are complex, data-intensive, and applicable only after students begin their studies. This study employs the Analytic Hierarchy Process (AHP) to create an indicator predicting academic performance for first-year students. This indicator combines commonly used variables with insights from past dropout cases, tailored to each university's and/or faculty context. The methodology was piloted with the Commercial Engineering department and the Technological Learning Complex at the University of Atacama in Chile. The index construction involved four stages: 1) defining dimensions and variables, 2) pairwise selection to determine weights, 3) index construction, and 4) validation. The index used 18 variables across four dimensions, with final weights assigned as follows: academic (49.7%), inclusion (32.8%), economic (8.9%), and social (8.6%). This academic risk index was tested against the 2018 cohort of the Commercial Engineering degree, showing an 82% match with actual student performance. This methodology offers a practical tool for creating academic performance indexes tailored to students entering to first time to the university, addressing the specific needs of individual institutions.