Articles liés à A Polytomous Nonlinear Mixed Model for Item Analysis:...

A Polytomous Nonlinear Mixed Model for Item Analysis: Item Calibration Using SAS PROC NLMIXED and PARSCALE - Couverture souple

Shin, Seon-Hi

 
9783639184310: A Polytomous Nonlinear Mixed Model for Item Analysis: Item Calibration Using SAS PROC NLMIXED and PARSCALE
Couverture souple
ISBN 10 : 3639184319 ISBN 13 : 9783639184310
Editeur : VDM Verlag, 2009

Synopsis

The study presented the formulation of an IRT model for polytomously scored Likert-type data under the framework of the nonlinear mixed model. The formulated polytomous nonlinear mixed model (PNLMM) was demonstrated to be equivalent to Muraki¿s rating scale model (1990). Due to the flexibility of the framework, the PNLMM was easily extended into a model for differential item functioning analysis and a regression-type model with a person-level covariate. Using simulations, the accuracy of SAS PROC NLMIXED in estimating the latent trait and the item parameters for the PNLMM was examined through eight common indicators. The correlations between PROC NLMIXED and PARSCALE estimates for model parameters were also evaluated. The results and their educational importance were discussed.

Les informations fournies dans la section « Synopsis » peuvent faire référence à une autre édition de ce titre.

Présentation de l'éditeur

The study presented the formulation of an IRT model for polytomously scored Likert-type data under the framework of the nonlinear mixed model. The formulated polytomous nonlinear mixed model (PNLMM) was demonstrated to be equivalent to Muraki¿s rating scale model (1990). Due to the flexibility of the framework, the PNLMM was easily extended into a model for differential item functioning analysis and a regression-type model with a person-level covariate. Using simulations, the accuracy of SAS PROC NLMIXED in estimating the latent trait and the item parameters for the PNLMM was examined through eight common indicators. The correlations between PROC NLMIXED and PARSCALE estimates for model parameters were also evaluated. The results and their educational importance were discussed.

Biographie de l'auteur

Seon-Hi Shin is an assistant professor in the Department of Advanced Studies in Education and Counseling at California State University, Long Beach. She teaches courses in statistics, psychometrics, and research methodology and conducts research activities related to her field. She received a Ph.D. from the University of Texas at Austin in 2003.

Les informations fournies dans la section « A propos du livre » peuvent faire référence à une autre édition de ce titre.

Éditeur
VDM Verlag
Date d'édition
2009
Langue
anglais
ISBN 10 
3639184319
ISBN 13 
9783639184310
Reliure
Broché
Nombre de pages
112
Coordonnées du fabricant
non disponible
Personne responsable
non disponible