Measuring Multidimensional Growth: A Higher-Order Item Response Theory Perspective

• In Event: Roundtable Session 18
  In Roundtable Session: Assessing Student Growth: Methods, Issues, and Applications

Abstract

Growth models have generally fallen into two camps: (1) longitudinal item response theory models; and (2) latent growth curve models. Psychometricians recently combined both models into a multilevel model with categorical outcomes, multiple latent traits at several time points, and individual growth parameters. The current study examines an extension of multilevel IRT growth to hierarchical IRT models using the SEM formulation. Conditions varied include correlation between latent traits, items loading on each dimension, and number of simulees. For each condition, item, person, and growth parameters are compared when using one of several model formulations (with or without the higher-order latent trait) or estimation algorithms (in a single stage or in two stages). Mplus code will be provided.

Authors

• Steven W Nydick, Pearson VUE

• Chun Wang, University of Minnesota

• Xinhui Xiong, CTB/McGraw-Hill LLC