Analyzing Item Response Data Using a Latent Space Modeling Approach in R

Half day short course (2:00pm – 5:30pm)

Short course #4

Intended Audience

This workshop introduces a novel latent space modeling framework for the analysis of item response data. In this approach, item response data are conceptualized as a bipartite network linking respondents and items, where a connection (or tie) is formed when a respondent answers an item correctly. A key advantage of this framework is its ability to generate an interaction map—a two-dimensional Euclidean representation—that visually captures relationships among respondents and items, as well as item–item and respondent–respondent associations.

The workshop will begin with an overview of latent space modeling for social network data and demonstrate how item response data can be understood as a special case of network data. I will then present the mathematical formulation of the latent space item response model along with its
Bayesian estimation procedure. Using empirical examples, participants will learn how to fit the model, process parameter estimates, and interpret the resulting interaction map.

All demonstrations will be conducted in R using a dedicated package developed for this latent space modeling approach to item response data.

About the Instructor

Minjeong Jeon (University of California, Los Angeles)

 Minjeong Jeon, Ph.D., is a Professor of Advanced Quantitative Methods in the School of Education and Information Studies at the University of California, Los Angeles (UCLA). She currently serves as Co-Editor-in-Chief of the Journal of Educational and Behavioral Statistics and as an Associate Editor for Psychometrika. She has recently been elected into the Editorial Council of Psychometrika.

Dr. Jeon’s research focuses on advancing statistical and computational methodologies for understanding complex social and behavioral data. Her work centers on the development, estimation, and application of latent variable models to study measurement, change, and individual differences over time. She has made significant contributions to areas such as latent space modeling, latent process and growth modeling, integrating modern statistical techniques to bridge theory, data, and real-world applications in education, psychology, and the social sciences.