Dropout Prediction and Reduction in Distance Education Courses with the Learning Analytics Multitrail Approach
Wagner Cambruzzi (Universidade do Vale do Rio dos Sinos - UNISINOS, Brazil)
Sandro José Rigo (Universidade do Vale do Rio dos Sinos - UNISINOS, Brazil)
Jorge L. V. Barbosa (Universidade do Vale do Rio dos Sinos - UNISINOS, Brazil)
Abstract: Distance Education courses are present in large number of educational institutions. Virtual Learning Environments development contributes to this wide adoption of Distance Education modality and allows new pedagogical methodologies. However, dropout rates observed in these courses are very expressive, both in public and private educational institutions. This paper presents a Learning Analytics system developed to deal with dropout problem in Distance Education courses on university-level education. Several complementary tools, allowing data visualization, dropout predictions, support to pedagogical actions and textual analysis, among others, are available in the system. The implementation of these tools is feasible due to the adoption of an approach called Multitrail to represent and manipulate data from several sources and formats. The obtained results from experiments carried out with courses in a Brazilian university show the dropout prediction with an average of 87% precision. A set of pedagogical actions concerning students among the higher probabilities of dropout was implemented and we observed average reduction of 11% in dropout rates.
Keywords: distance education, learning analytics
Categories: L.3, L.3.5, L.3.6