Model-Based Approaches to Classifying Students Into Competence-Based Proficiency Levels

• In Event: AERA Virtual Poster Session 8
  In Poster Session: Poster Session 1

Abstract

The aim of this study is to investigate and compare performances of three model-based methods, i.e., rating scale model (RSM) combined with Wright-Map, latent class model (LCM), and mixture rating scale model (MRSM), on determining the proficiency levels of students on a competence-based scale, using both simulation and empirical study. Preliminary results showed that MSRM performs best in general among three methods across limited conditions of the simulation study, which justifies a full-scale study for other conditions of the simulation study and the empirical case. Significance and implication of the study were also discussed.

Authors

• Jing Huang, East China Normal University

• Xiangdong Yang, East China Normal University