A hybrid modeling approach for predicting the educational use of mobile cloud computing services in higher education

Abstract

The decision to integrate mobile cloud computing (MCC) in education without determining optimal use scenarios is a universal problem as the adoption of such services becomes widespread. Accordingly, this study developed and validated a predictive model that explains the role of students' information management (i.e. retrieve, store, share, and apply) practices in predicting their attitudes toward using the MCC services for educational purposes. This study validated the model by the complementary use of machine learning algorithms alongside a classical SEM-based approach based on data collected from 308 undergraduate students. The SEM results indicated that the students' information management (i.e. retrieve, store, share, and apply) practices were significantly associated with their attitudes, which were significantly associated with the behavioral intentions. The structural model explained a significant portion of the variance in the behavioral intentions. Likewise, the classifier model suggested that the students’ information management practices and attitudes predicted the behavioral intentions. Further, the applied algorithms predicted the behavioral intentions with an accuracy of more than 72% in most cases. Thereby, the study extended an original theory (TRA) into the MCC area by using a multi-analytical approach. The findings implied that employing the MCC services for personal information management should be supported and encouraged in the higher education by designing authentic learning environments and scaffolding the students in using such services.