From Face-To-Face To Virtual Education. An Evolutionary Analysis Of The Academic Performance Of First-Level Students Of The Universidad Politécnica Estatal Del Carchi Through Neural Networks

Abstract

Prior to the confinement caused by covid19, the face-to-face modality covered a series of aspects that were related to the academic performance of students and that, because of the migration to virtual education, gave way to other variables that now constitute the new causes of their success or failure. This research develops a longitudinal analysis of the evolution of the academic performance of students in the first levels of the Universidad Politécnica Estatal del Carchi, through a comparative statistical methodology based on neural networks that takes as moments of study two academic periods before and two after the pandemic. The research yielded a predictive model that offers a 78% of correct classification in relation to the main variables that can influence academic performance, being results that coincide to a great extent with the research taken as references.