ABSTRACT
Adaptive learning systems can assist in effectively teaching learners with a wide range of perceptual and processing variability, but their effectiveness remains tied to material designed into them. Given increasing enrollment of college students with disabilities, an approach such as Universal Design for Learning (UDL) can guide making course material accessible in beneficial ways for all students. However, implementing UDL and adaptive features takes time, often through iterative (re)design. Thus, educators face helping struggling students when a course is not (yet) fully universally designed. This study presents a prescriptive analytics approach utilizing formative data traces gathered in a data warehouse. It compares simulated worlds in a Bayesian network to identify points during a course where recommending tutoring may be warranted to support struggling students. Data come from a high enrollment, accelerated, introductory online undergraduate English course at a women's institution utilizing both adaptive learning and online tutoring with human tutors. This approach could be employed in other courses or institutions to indicate where it is not currently clear how to make course content work smoothly for all students from a standalone universal design standpoint (or applying sufficient resources remains needed) and thus where individualized tutoring is predicted to be beneficial.
