Using Blackboard Learn data to build predictive models of student success is an active area of learning analytics practice with success for specific courses. However, it is less clear what pedagogical practices contribute to a model that can be used across all courses at a campus. In this presentation, we will discuss the data points that have proven most powerful at multiple campuses, and how you can use Learn to create this data. We will look at the results through multiple learning analytics applications, including Analytics for Learn, Blackboard Predict, and homegrown solutions. By adopting these uses, it is possible to turn the LMS into a highly effective tool for predicting student performance.
Presenters: Venugopal Ambadipudi, John Fritz