You are cordially invited to join our second seminar in the ACLA Brown Bag Seminar Series, with our very first international speaker: Benjamin Deonovic.
Few models have been more ubiquitous in their respective fields than Bayesian knowledge tracing and item response theory. Both these models were developed to analyze data on learners. However, the study designs that these models are designed for differ; Bayesian knowledge tracing is designed to analyze longitudinal data while item response theory is built for cross-sectional data. This paper illustrates a fundamental connection between these two models. Specifically, the stationary distribution of the latent variable and the observed response variable in Bayesian knowledge tracing are related to an item response theory model. This connection between these two models highlights a key missing component: the role of education in these models. A research agenda is outlined which answers how to move forward with modeling learner data.
Benjamin Deonovic is a research scientist at ACTNext. He earned his PhD and MS in Biostatistics from the University of Iowa and his BS in bioinformatics, computational biology, and mathematics from Iowa State University. His thesis research at the University of Iowa focused on exploring the use of binary variables as parameters in statistical models in the Bayesian framework. Benjamin was drawn to ACT by its mission to help people achieve education and workplace success. Toward achieving that goal, he seeks to apply his expertise in machine learning, Bayesian statistics, and computation to problems in psychometrics, educational measurement, and diagnostic measurement. His research focuses on connecting statistical models for learning and assessment, developing innovative psychometric models to incorporate process data, and writing efficient procedures for computationally complex tasks.