february, 2023
Event Details
The talk discusses an exploratory graphical modeling approach for evaluating local item dependence in cognitively diagnostic models (CDMs). CDMs are restricted latent class models that describe an individual’s status on a set of discrete latent attributes (e.g., mastery of cognitive skills, pathological symptoms). In this talk, I present an exploratory graphical modeling approach to evaluating local item dependence within the CDMs. The model integrates a pairwise Markov random field to the generalized CDM so that items’ interdependence can be modeled via an undirected spatial network.
Speaker Bio: Hyeon-Ah Kang is an assistant professor in the Quantitative Methods program in the Department of Educational Psychology. She also serves as an associate director of the Center for Applied Psychometric Research at the
College of Education. She studies theoretical and applied statistics in educational and psychological measurement.
