2022 Dissertation Prize
Restricted latent class models (RLCMs) provide a pivotal framework for supporting diagnostic research that enhances human development and opportunities. In earlier research, the focus was on confirmatory methods that required a pre-specified expert-attribute mapping known as a Q matrix. Recent research directions have led to the creation of exploratory methodology that is able to infer the Q matrix without expert intervention. Within this talk, we focus on novel Bayesian formulation of a less restrictive monotonicity condition when estimating the underlying latent structure and attributes. Moreover, we extend the framework to allow for using logit-link function instead of the probit and addressing the dependency structure found among attributes with a higher-order model that generalizes under an exploratory factor analysis (EFA).
About the Speaker
James Balamuta is a Visiting Assistant Professor in the Department of Statistics at the University of Illinois Urbana-Champaign. He holds a Ph.D. in Informatics from the University of Illinois Urbana-Champaign. His research focuses on latent variable estimation with a specialization in restricted latent class models, which provide considerable insight in psychological and educational assessment. He also has a strong interest in data science pedagogy and figuring out innovative ways to incorporate industry data science techniques in academic settings.