A differential item functioning analysis of the PSDQ with Turkish and New Zealand/Australian adolescents

Abstract

Objective: The use of item response theory models (IRT) in sport and exercise psychology has provided some unique insights into the quality of measure especially at the item level Fletcher. R. B. (August 2005). Detecting item bias in physical self-concept: an application of differential bundle functioning. In I Morris, R Terry, S. Gordon, S. Hanrahan, L. Ievleva. G. Kolt, & P. Tremayne (Eds.), Promoting health and performance for life: Proceedings of the ISSP 11(th) world congress of sport psychology [CD-ROM]. Sydney: International Society of Sport Psychology; Fletcher, R. B., & Hattie, J. A. (2004). An examination of the psychometric properties of the Physical Self-Description Questionnaire using a polytomous item response model. Psychology of Sport and Exercise, 5, 423-446; Fletcher, R. B., & Hattie. J. A. (2005). Understanding gender differences in physical self-concept: an example using multidimensional differential item functioning. Educational and Psychological Measurement, 65, 657-667.]. IRT approaches using differential item functioning (DIF) provide a powerful method to examine the equivalence of item responses across different cultures. Recent advances in DIF methodology. polytomous DIF (PDIF), allow for a more in depth analysis of how items differ between groups at the response option and item level. Method: From a cross-cultural perspective this paper demonstrates the advantages of applying Shealy and Stout's [(1993). A model-based standardization approach that separates true bias/DIF from group ability differences and detects test bias/DTF as well as item bias/DIF. Psychometrika, 85, 159-194.] polytomous simultaneous item bias test (Poly SIBTEST) to the Physical Self-Description Questionnaire (PSDQ) [Marsh, H. W., Richards, G. E., Johnson, S., Roche, L., & Tremayne, R (1994). Physical Self-Description Questionnaire: psychometric properties and a multitrait-multimethod analysis of relations with existing instruments. Journal of Sport and Exercise Psychology, 15, 270-305.], using item responses from Turkish and New Zealand/Australian (NZ/A) adolescents. Results: The main findings of this research showed that items biased in favor of New Zealand males and females were more self-referenced, whereas items that showed DIF in favor Turkish males and females were from an external frame of reference. (C) 2008 Elsevier Ltd. All rights reserved.