Interesting article that studies the reasons for persistently high dropout rates in online learning. It's a largish studies, with the authors arguing that the telephone survey size of 226 participants has a 99% chance of detecting any 'theme' that has a 5% incidence or greater in the general population. I would suggest, however, that the quantifications of those themes are less reliable. Still, the survey points to personal circumstances (moved, changed jobs, etc) and learner context (time management, resources) as being major factors, with institutional and teaching context (admin, design and delivery) being far down the list. Existing interventions (surveyed near the top of the article) order these in reverse sequence, which may explain why they have been ineffective.