The longitudinal association between engagement and achievement varies by time, students’ profiles, and achievement state: A full program study

There is a paucity of longitudinal studies in online learning across courses or throughout programs. Our study intends to add to this emerging body of research by analyzing the longitudinal trajectories of interaction between student engagement and achievement over a full four-year program. We use learning analytics and life-course methods to study how achievement and engagement are intertwined and how such relationship evolves over a full program for 106 students. Our findings have indicated that the association between engagement and achievement varies between students and progresses differently between such groups over time. Our results showed that online engagement at any single time-point is not a consistent indicator for high achievement. It takes more than a single point of time to reliably forecast high achievement throughout the program. Longitudinal high grades, or longitudinal high levels of engagement (either separately or combined) were indicators of a stable academic trajectory in which students remained engaged —at least on average— and had a higher level of achievement. On the other hand, disengagement at any time point was consistently associated with lower achievement among low-engaged students. Improving to a higher level of engagement was associated with —at least— acceptable achievement levels and rare dropouts. Lack of improvement or “catching up” may be a more ominous sign that should be proactively addressed. •The relationship between engagement and achievement is complex, longitudinal, and heterogenous.•Only a subset of students maintained stable and consistent engagement with high academic achievement.•Students who improved after a turbulent start were able to maintain a reasonable level of achievement.•Students who disengaged or failed to improve their engagement were at the highest risk of low achievement.•A method for multi-channel sequence mining is described for studying states, transitions, and trajectories.