An IRT model for multidimensional ranking data in ipsative tests

Xuelan QIU, Wen Chung WANG, Shungwon RO

Research output: Contribution to conferencePaper

Abstract

By design, multidimensional raking data are ipsative and cannot measure the absolute level of latent traits. In this study, we developed an IRT model for multidimensional ranking data where different statements measure different latent traits. We conducted a brief simulation study to evaluate parameter recovery and provided an empirical example.
Original languageEnglish
Publication statusPublished - Apr 2015

    Fingerprint

Citation

Qiu, X.-L., Wang, W.-C., & Ro, S. (2015, April). An IRT model for multidimensional ranking data in ipsative tests. Paper presented at The NCME 2015: 77th Annual Meeting of National Council on Measurement in Education, InterContinental Hotel, Chicago, Illinois.