web stats
 
   
 
Selected Research Project
 
Project Title Person–item Interaction in Likert Items: Model Development and Operational Issues
 
Principal Investigator Professor Wang Wen Chung
 
Area of Research Project
Educational Measurement and Assessment
 
Project Period
From 1/2010 To 12/2011
Objectives
  1. To invent a new IRT-based unfolding model to account for person–item interaction in Likert items, root it in a multi-level framework, and demonstrate its advantages over traditional models through a series of simulations;
  2. To demonstrate implication and applications of the new model through real data analyses of Likert items in the 2003 TIMSS and the 2006 PIRLS; and
  3. To adapt test linking procedures, item selection rules, and control procedures of item exposure and test overlap in CAT (computerized adaptive testing) and CCT (computerized classification testing) for use with the new technique, and evaluate their performances under a variety of conditions.
Methods Used
  • Development of a new model for Likert items; ƒÜ
  • Development of parameter estimation procedures and computer programs; ƒÜ
  • Simulation studies; ƒÜ
  • Real data analyses; and ƒÜ
  • CAT & CCT implementation.
Summary of Findings

The new model has been validated. Its advantages over standard IRT-based unfolding models are demonstrated. Operational issues, including test linking, CAT, and ACT are thoroughly investigated to ensure that the new technique is ready for use.

Impact
  • Person–item interaction in Likert items can be analyzed more properly with the new model, such that test validity and reliability can be better assessed, and test fairness can be better ensured. ƒÜ
  • Guidelines, computer programs, and annotated examples can be provided to facilitate the use of the new technique in many areas.
Selected Publications Related to the Study
  • Wang, W.-C., & Liu, C.-W. (2011). Computerized classification testing under the generalized graded unfolding model. Educational and Psychological Measurement, 71, 114–128.
  • Wu, S.-L., & Wang, W.-C. (2011, July). The confirmatory multidimensional generalized graded unfolding model. Paper presented at the 76th Annual and the 17th International Meeting of the Psychometric Society, Hong Kong.
  • Liu, C.-W., & Wang, W.-C. (2011, July). The multilevel generalized graded unfolding model. Paper presented at the 76th Annual and the 17th International Meeting of the Psychometric Society, Hong Kong.
  • Liu, C.-W., & Wang, W.-C. (2011, December). The facets IRT unfolding model. Paper presented at the Annual Meeting of TERA International Conference on Education, Taiwan.
Biography of Principal Investigator
Professor Wang Wen Chung is the Director of the Assessment Research Centre and Chair Professor of Educational and Psychological Measurement. Prior to joining HKIEd in 2008, he was a Distinguished Professor and Head of the Department of Psychology at National Chung Cheng University in Taiwan. He obtained his Ph.D. from the University of California, Berkeley. He was a Fulbright grantee in 2005, and a visiting professor at UC Berkeley in 2005 and 2007. He has received Research Excellence Awards from the National Science Council and National Chung Cheng University, as well as the Mu-Dou Award (¤ìÅM¼ú) for his contribution to education. He has published more than 140 refereed journal articles, in addition to several books and chapters. He is an editorial board member of several journals and a reviewer of many journals. Since joining HKIED in 2008, he has received (as principal investigator) three research grants from the General Research Fund and one research grant from the Public Policy Research Fund.
Funding Source

General Research Fund