Meaningful Online Interactions as a Predictor of Student Learning Outcomes in Online Learning Environments: Moderating Effect of Student Differences

Abstract

The purpose of this study was to investigate the direct influence of meaningful interactions, as classified by Moore’s interaction model, on student learning outcomes in online learning environments. Specifically, this study explored the predictive relationships between three types of online interactions (learner-instructor interaction (LII), learner-learner interaction (LLI), and learner-content interaction (LCI)) and student learning outcomes, namely student satisfaction and perceived learning in online courses, focusing on their unique and combined contributions. Furthermore, the study examined the moderating effects of student-related variables (gender, prior online learning experience, and academic degree level) on these relationships. The study involved 217 undergraduate and graduate students from the College of Education at Ohio University who had previously enrolled in at least one online course. Data was collected during the Spring semester of 2024. A quantitative, correlational research design was employed. A self-reported survey instrument incorporating validated scales for student satisfaction, perceived learning, and the three types of interactions (learner-instructor interaction (LII), learner-learner interaction (LLI), and learner-content interaction (LCI)) within an online course was administered. A pilot test was conducted to assess the survey's validity and reliability and research procedures, showing strong measurement consistency within the sample. Data analysis utilized multiple regression to examine predictive relationships among the variables. Moderated multiple regression, implemented through Hayes' PROCESS macro and hierarchical regression, was employed to investigate the potential moderating effects of student-related variables on the relationship between interaction types and student outcomes. Results indicated that the three types of interaction (LCI, LII, and LLI) significantly predict SAT and PL in online courses. Among the predictors, LCI demonstrated the strongest and most significant predictor of both student satisfaction and perceived learning, followed by LII, which had a significant impact on SAT but a limited effect on PL. LLI demonstrated a moderate contribution to PL but had a minimal impact on SAT. Furthermore, the moderation analysis revealed that academic degree level and prior online learning experience significantly moderated the relationship between LCI and PL. In contrast, gender did not significantly moderate the relationship between the examined interaction types (LCI and LLI) and perceived learning. Additionally, none of the examined moderators influenced the relationship between interaction types (LCI and LII) and student satisfaction.