Exploring synchronous, asynchronous, and conventional online courses in higher education

Online learning has expanded substantially in recent years, yet comparatively little is known about how different online modalities influence learners' perceptions and their intention to continue learning. This study compared three predominant online learning modalities-synchronous (Syn), asynchronous (Asyn), and conventional online courses (COC)-using an extended Information Systems Success Model (ISSM) integrating system quality (SQ), information quality (IQ), teaching presence (TP), self-efficacy (SE), perceived usefulness (PU), learning satisfaction (LS), and continuance intention (CI). The study addressed two questions: (1) how the latent means of ISSM constructs differ across Syn, Asyn, and COC, and (2) whether the structural paths among these variables differ across modalities. Survey data from 795 undergraduate students in South Korea were analyzed using latent mean analysis and multi-group structural equation modelling. Results showed that COC consistently demonstrated higher latent mean values than Syn and Asyn, underscoring the benefit of systematically designed courses. Multi-group analysis further revealed significant modality-specific variations, with relationships such as IQ → LS and PU → CI being strongest in the COC model. Overall, the findings indicate that Syn, Asyn, and COC engage ISSM mechanisms in distinct ways, highlighting the importance of modality-sensitive instructional and system design. These insights provide evidence-based guidance for enhancing the quality, satisfaction, and sustainability of online learning environments.