On the Interpretation of Sequential Models: A Multilevel Modeling Approach

• In Event: AERA Poster Session 6
  In Poster Session: Division D Section 1 Poster Session 2

Abstract

To increase model flexibility in the investigation of different response styles, the IRTree models have been proposed, in which a tree structure is used to represent the outcomes of item response processes. Despite the increasing popularity of the sequential models, questions regarding how to interpret the result and how to compare pseudo-items from sequential models remain unanswered. This project uses the multilevel model approach to analyze sequential models by using the nest nature of pseudo-items. We argue that sequential models should be interpreted sequentially so that pseudo-items can be compared within a test.

Authors

• Yibing Wang, Shandong University of Finance and Economics

• Jiaqi Zhang, SKT Education Group