Time Really Matters: Understanding the Temporal Dimension of Online Learning Using Educational Data Mining

Learning and participation are inseparable in online environments. To improve online learning, much effort has been devoted to encouraging online participation. However, previous research has investigated participation from a variable-based perspective, looking only for relationships between participation and other variables. Time can change and even reverse this relationship, however. Online learning is a cumulative process, but participation at several critical moments is more significant for learning than is participation at other points. To fully understand how learning unfolds over time, it is necessary to shift to a new perspective on learning. Adopting an event-based view on which the units of analysis are separate but interrelated learning events, this study investigates longitudinal patterns in online participation. Using data mining techniques for education, the study validated longitudinal patterns of participation as an accurate measure for differentiating learner performance. In addition, the first segment of this online learning experience was identified as the most critical moment in which educators should provide efficient interventions to help their learners maintain active participation. Additional design, pedagogical, and methodological implications for online teaching and facilitation practices are discussed at the end.