Data Analytics on Online Student Engagement Data for Academic Performance Modeling
Xiaohui Tao, Aaron Shannon-Honson, Patrick Delaney, Lin Li, Christopher Dann, Yan Li, Haoran Xie
Abstract
In large MOOC cohorts, the sheer variance and volume of discussion forum posts can make it difficult to for instructors to distinguish nuanced behaviour in students such as positive engagement or stress. Sentiment analysis has been used to build student behavioural models, however, more recent research suggests that separating sentiment and stress into different measures could improve text analysis in this domain. Detecting stress in a MOOC corpus is challenging as students may use language that does not conform to standard definitions, but new techniques like TensiStrength provide more nuanced measures of stress. In this work, we introduce an ensemble method that extracts features of engagement, semantics and sentiment and stress from an AdelaideX student dataset. Stacked and voting methods are used to compare performance measures on how accurately these features can predict student grades. The stacked method performed best across all measures, with our Random Forest baseline further demonstrating that negative sentiment and stress had little impact on academic results. As a secondary analysis, we explored whether stress among student posts increased in 2020 compared to 2019 due to COVID-19 to understand the impact of major events on online learners, but found no significant change. Importantly, our model indicates that there may be a relationship between features, which warrants future research.