Parameter Recovery from Higher Order Item Response Theory Models with Structural Missingness

• In Event: Missing Data

Abstract

By combining HO-IRT models with structural missingness researchers can control instrument length while still measuring hierarchical general and subdomain constructs. This study provides insight into the application of HOIRT models with structural missingness by exploring how group design, sample size, and correlation influence parameter recovery.

Authors

• Yale Quan, University of Washington

• Chun Wang, University of Washington