A two-staged SEM: artificial neural network approach for understanding and predicting the factors of students’ satisfaction with emergency remote teaching

Abstract

This study seeks to address knowledge gaps regarding the role of self-regulated learning as a mediator in the relationship between interactions, internet self-efficacy, and student satisfaction. We conducted a survey of 1590 students from north Indian universities about their level of satisfaction, self-regulated learning, internet self-efficacy, and different interactions (learner-learner interaction, learner-content interaction, and learner-instructor interaction) during emergency remote teaching. By employing a two-stage SEM-ANN approach, this study contributes to methodological advancements and provides a comprehensive analysis of complex relationships. According to the findings, the identified factors are significant predictors of students’ satisfaction with online education in synchronous settings. Our research also shows that self-regulated learning fully mediates the effect of internet self-efficacy on student satisfaction during emergency remote teaching. This suggests that internet self-efficacy alone may not guarantee student satisfaction unless accompanied by self-regulated learning skills.