The Standardized Mean Difference Within the Framework of Item Response Theory.

A study was conducted to demonstrate how the standardized mean difference can be generalized into an item response theory (IRT)-based effect size measures. A disattenuation procedure based on the IRT test reliability was suggested to rectify the attenuation on the sampling variance due to the measu...

Descripción completa

Detalles Bibliográficos
Publicado en:Educational & Psychological Measurement Vol. 64; no. 2; pp. 201 - 224
Autores principales: Wang, Wen-Chung, Chen, Hsueh-Chu
Formato: Artículo
Publicado: Sage Publications Inc. Apr2004
Materias:
Acceso en línea:Ver este registro en EBSCOhost
Descripción
Sumario:A study was conducted to demonstrate how the standardized mean difference can be generalized into an item response theory (IRT)-based effect size measures. A disattenuation procedure based on the IRT test reliability was suggested to rectify the attenuation on the sampling variance due to the measurement error. Findings revealed that the direct generalization of the standardized mean difference into the IRT context was applicable and the disattenuation procedure was appropriate. Furthermore, an empirical example concerning English accents for Taiwan students was outlined as an illustration of the implication and applications of the proposed method.