The Influence of Educational and Emotional Support on e-Learning Continuance Intention: A Two-Stage Approach PLS-SEM and ANN Analysis

Abstract

In the context of the digital era, e-learning has become an innovative and indispensable component of the educational sector. With the continuous advancement of technology and the widespread adoption of the internet, e-learning has demonstrated its key role in maintaining educational continuity and supporting remote teaching. However, despite the extensive applications and significant advantages of e-learning, the willingness of college students to continue using e-learning platforms is not always high, presenting a challenge for educators and technology developers. Based on the Expectation Confirmation Model, this study examines the influence of perceived educational and emotional support on the continuance intention of e-learning among college students. The researchers conducted a survey using a structured questionnaire randomly among 379 university students from three universities in Henan Province to measure their self-reported responses on six constructs: perceived educational support, perceived emotional support, perceived usefulness, confirmation, satisfaction, and continuance intention. The study uses the Structural Equation Modeling—Artificial Neural Network (SEM-ANN) method to elucidate the non-compensatory and non-linear relationships between predictive factors and e-learning continuance intention. Except for the direct effects of perceived educational support and perceived emotional support on perceived usefulness, which were not significant, all other hypotheses were confirmed. Moreover, according to the normalized importance obtained from the multilayer perceptron, satisfaction (100%) was found to be the most critical predictive factor, followed by confirmation (29.8%), perceived usefulness (28.2%), perceived educational support (22.7%), and perceived emotional support (21.8%). All constructs together accounted for 62.0% of the total variance in college students’ e-learning continuance intention. This study’s adoption of a two-stage analysis approach improved the depth and accuracy of data processing and expanded the methodological scope of educational technology research. It provides direction for future in-depth studies in various environments and cultural contexts.