Using Stan for Bayesian Psychometric Modeling (Part I)

Session Type: Training Session

Abstract

This session will provide attendees with systematic training on Bayesian estimation of classic psychometric models as well as newly developed models using Stan, with a particular focus on helping graduate students who are searching for dissertation topics to navigate the vast body of Bayesian psychometric literature. The estimation of model parameters for common and sophisticated psychometric models will be illustrated and demonstrated using Stan. Although this workshop places a particular emphasis on IRT models, other psychometric models such as generalizability theory, classic test theory, confirmatory factor analysis, latent class models, cognitive diagnostic models, and structural equation models will also be covered. Further, the advantages and disadvantages of Stan compared to traditional Bayesian software programs such as OpenBUGS and JAGS will be discussed.

This session consists of lecture, demonstration, and hands-on activities of running Stan. It is intended for intermediate and advanced graduate students, researchers, and practitioners who are interested in learning the basics and advanced topics related to parameter estimation of psychometric models using Stan.

Sub Unit

• Training Sessions

Session Organizer

• Yong Luo, ETS

Presenter

• xin qiao, Southern Methodist University