Applying Probabilistic Models for Knowledge Diagnosis and Educational Game Design
Anna Rafferty

About the research


NAEd/Spencer Dissertation Fellowship

Award Year



University of California, Berkeley

Primary Discipline

My research focuses on creating formal models of teaching and learning, and applying these probabilistic models to assess student knowledge and make pedagogical decisions. One part of my research involves automatically assessing student knowledge via observed behavior in complex interactive environments, such as virtual labs and games. I have validated this approach in lab experiments, and I am now using it to assess students' algebra knowledge and to diagnose student misunderstandings via STEM games. This model can recognize the source of errors, such as conceptual versus arithmetic errors, and does not interrupt student work. Its assessments can be provided to teachers or to a computer tutor to personalize instruction, which is the focus of the second part of my dissertation. Applying a formal modeling approach to these problems allows education to benefit from machine learning and statistics and allows one to combine ideas from different strands of educational technology research.
About Anna Rafferty

Pin It on Pinterest