Investigating Responses and Missingness Using Item Response Tree Models: Illustration With Empirical Data

• In Event: 72.066 - Poster Session 16
  In Poster Session: 72.066-7 - Division D Section 2 Poster Session

Abstract

An item response tree model makes it possible to estimate item parameters and person parameters when there are missing responses in a data set, without manipulating the data. This model creates new response categories for missing responses in addition to correct and incorrect responses or other categorical responses. By using the item response tree model, a person’s proficiency, tendency to omit items, and the correlation between the two are measured and estimated. We explore assessment data using the model to improve understanding of the data with missing responses.

Authors

• Seohyun Kim, University of Georgia

• Zhenqiu Lu, University of Georgia

• Allan S. Cohen, University of Georgia