Fitting an Item Response Theory–Latent Growth Curve Model in a Single-Group Anchor Test Design

• In Event: 70.066 - Roundtable Session 28
  In Roundtable Session: 70.066-6 - Applications of Advanced Item Response Theory Models

Abstract

Psychometricians recently proposed to combine an IRT model and a latent growth curve model into a multilevel model to reliably quantify individual growth. However, the existing research only used real data to show the application of the IRT-LGC model. In this proposal, we propose a thorough simulation study to evaluate the performance of IRT-LGC in a single-group anchor test design with respect to model convergence and parameter recovery. Two types of IRT-LGC models are considered, one for the same-anchor design, and the other for the different-anchor design. Both the robust maximum likelihood estimation and the weighted least squares estimation implemented Mplus will be used. The full set of results will be provided upon completion of the project.

Authors

• Chun Wang, University of Minnesota

• Nidhi Kohli, University of Minnesota - Twin Cities