E-Learning Satisfaction During the Covid-19 Pandemic: A Non-Parametric Approach

Abstract

This paper aims to explore the relationships between e-learning dimensions and students’ perceived satisfaction in the context of a public university in a developing country during the time of coronavirus (COVID-19) pandemic. Survey questionnaire forms designed in 7-point Likert scale were completed by a total of 133 business undergraduates who had completed purely on-line courses as their university campuses were closed to prevent from coronavirus spread. A non-parametric approach, i.e., Partial Least Squares – Structural Equation Modelling (PLS-SEM) was used to test the e-learning model.  A measurement model and a structural model of the specification was analyzed to test internal reliability, indicator reliability, discriminant validity, convergent validity, coefficient of determination (R-squared), predictive relevance, path coefficients and effect size. Key analysis results have showed that the most critical factor positively affecting student satisfaction with e-learning is learner interaction while computer anxiety is negatively associated with e-learner satisfaction with this learning mode. Content analysis has also indicated that interaction is one of the most common problems as perceived by e-learners. The results emphasized the importance of human interaction in virtual learning environment. E-learning system designers should consider more functions for easier and more dynamic interactions in online class sessions.