Abstract
The volume of data that can be captured and stored from students' everyday interactions with digital environments allows for the creation of models of student knowledge, skills, and attributes unobtrusively. However, models and techniques for transforming these data into information that is useful for educators have not been established. This paper explores the use of learning progressions and Bayesian networks (BayesNets) as tools for aggregating evidence across digital learning environments. It includes a worked example demonstrating both the technical creation of a BayesNet and potential classroom application to monitor students' proficiency as they interact with a range of digital media environments.
Notes on contributor
Kristen DiCerbo is a Principal Research Scientist at Pearson. Her research program centers on the use of simulations and games for learning and assessment, as well as communication of results for instructional decision making.