Considering the Ideal Point Response Process in Large-Scale Educational Assessments: An Unfolding Tree Model

• In Event: Measurement and Statistical Modeling

Abstract

This study attempted to construct an Unfolding Tree model based on the IRTree model and the ideal point response process, and verified the effectiveness of this model through simulation and empirical (National Assessment of Education Quality of China) data. We found that the Unfolding Tree model can fit the data of attitudinal Likert scale better than the IRTree model; the mixture Unfolding Tree model can adapt to a wider range of measurement situations and has better fit performance than the single Unfolding Tree model. Also, the mixture Unfolding Tree model can realize the accurate classification of potential trait categories of participants. The unfolding Tree model can provide a model basis for background questionnaires analysis in the large-scale education evaluation program.

Authors

• Biao Zeng, Beijing Normal University - Beijing

• HongBo Wen, Beijing Normal University Collaborative Innovation Center of Assessment toward Basic Education Qual…

• Junjie Zhang, Beijing Normal University