Collapsing or not? A Practical Guide to Handling Sparse Responses for Polytomous Items (Poster 16): Division D - Section 1: Measurement and Research Methodologies, 11:37 AM

• In Event: e-Lightening Ed-Talk Session 2

Abstract

In ordinal data analysis category collapse is the process of combining adjacent response options to create fewer response categories than were originally measured. When collapsing response categories researchers need to be aware of inducing data-model misfit and of obtaining biased parameter estimates. Through simulation studies we found that using the Graded Response and Generalized Partial Credit Item Response Theory Models with items containing non-central collapsed categories does not induce significant data-model misfit. Furthermore, we show that person parameters from both models are well recovered. Item parameters recovered from the Generalized Partial Credit model show significant bias when fit to an item containing collapsed categories.

Authors

• Yale Quan, University of Washington

• Chun Wang, University of Washington