Virtual Education: Impact of Socio-Emotional and Pedagogical Factors on Academic Performance Based on Neural Networks and Stepwise Regression

EasyChair Preprint 6289

Abstract

The objective was to determine the impact of socio-emotional and pedagogical factors on the academic performance of students from the UNAM professional school of Management and Social Development during the COVID 19 pandemic, based on neural networks and stepwise regression. From the 38 indicators of socio-emotional and pedagogical factors applied to 236 students, the stepwise regression model identified 7 significant indicators (pedagogical and personal problems) that impact on academic performance, the neural network model determined the importance of indicators that influence academic performance. academic performance, considering the personal problems first, followed by pedagogical and family ones. The mean square error for the stepwise regression model is 7.25 and the coefficient of determination was 0.221, for the neural network model the mean square error was 6.78, while the coefficient of determination was 0.256. According to the results of both methodologies, the students of the professional school of Management and Social Development showed pedagogical and personal problems in their academic performance during the COVID 19 pandemic.

Keyphrases: Academic, COVID-19, Pandemic, Performance, neural network, stepwise regression

@booklet{EasyChair:6289,
  author    = {José Luis Morales Rocha and Mario Aurelio Coyla Zela and Nakaday Irazema Vargas Torres and Jarol Teófilo Ramos Rojas and Santos Octavio Morillos Valderrama and Genciana Serruto Medina},
  title     = {Virtual Education: Impact of Socio-Emotional and Pedagogical Factors on Academic Performance Based on Neural Networks and Stepwise Regression},
  howpublished = {EasyChair Preprint 6289},
  year      = {EasyChair, 2021}}