Student’s behavior in an online course is strongly influenced by the learning design of the course, however there are not many discipline-focused studies investigating how students engage with the Learning Management System and course materials. This paper investigates an asynchronous and fully online introductory statistics course with a collaboration component (Collaborative Keys). This study evaluates how students use these and other resources, if they are being used as intended, and how the use is related to performance. We extend the educational analytics literature and introduce a new measure to quantify how students transition between resources, and we evaluate how these transitions differ by student performance over the course of an entire term. The results suggest that the use of course’s resources is related to student achievement, with higher performing students focusing on video-related resources and showing more consistency and effectiveness in their use of resources throughout the term.

In this paper, we discuss generalized mixture models and related semi-supervised learning methods, and show how they can be used to provide explicit methods for unknown class inference. After a brief description of standard mixture modeling and current model-based semi-supervised learning methods, we provide the generalization and discuss its computational implementation using three-stage expectation-maximization algorithm. © Springer-Verlag 2007.