In this talk, I will review examples from a program of research that, over the last 15+ years, led to a significantly broadened set of parametric item response theory models, estimation algorithms, methods for model fit appraisal, and software implementation. Applications of methods from this program of research will be discussed. It is argued that these methods are now sufficiently developed for routine use in educational and psychological measurement research and practice.
about the speaker
Li Cai is a Professor of Education and Psychology at UCLA, and a Managing Partner at Vector Psychometric Group, LLC. He earned his PhD in Quantitative Psychology from the historic LL Thurstone Psychometric Laboratory at University of North Carolina – Chapel Hill. Broadly speaking, his research agenda involves the development, integration, and application of latent variable models in educational, psychological, and clinical outcomes measurement. One major component on this agenda is statistical computing, particularly as related to IRT and multilevel modeling. As director of UCLA’s storied educational assessment research center CRESST, he has managed a number of large-scale K-12 assessment development and sustainability projects. He has been an editor or associate editor of major psychometric and statistical methodology research journals and has served on governing bodies of several professional associations in the field.