作者: Dragan Gašević , Shane Dawson , Tim Rogers , Danijela Gasevic
DOI: 10.1016/J.IHEDUC.2015.10.002
关键词: Knowledge management 、 Learning Management 、 Learning analytics 、 Estimation 、 Self-regulated learning 、 Computer science 、 Predictive power 、 Blended learning
摘要: Abstract This study examined the extent to which instructional conditions influence prediction of academic success in nine undergraduate courses offered a blended learning model (n = 4134). The illustrates differences predictive power and significant predictors between course-specific models generalized models. results suggest that it is imperative for analytics research account diverse ways technology adopted applied contexts. use, especially those related whether how learners use management system, require consideration before log-data can be merged create predicting success. A lack attention lead an over or under estimation effects LMS features on students' These findings have broader implications institutions seeking portable identifying students at risk failure.