Using Loss Functions for DIF Detection: An Empirical Bayes Approach.

Zwick, Rebecca; Thayer, Dorothy; Lewis, Charles

Notes FAQ Contact Us

Back to results

Peer reviewed

ERIC Number: EJ615871

Record Type: Journal

Publication Date: 2000

Pages: N/A

Abstractor: N/A

ISBN: N/A

ISSN: ISSN-1076-9986

EISSN: N/A

Available Date: N/A

Using Loss Functions for DIF Detection: An Empirical Bayes Approach.

Zwick, Rebecca; Thayer, Dorothy; Lewis, Charles

Journal of Educational and Behavioral Statistics, v25 n2 p225-47 Sum 2000

Studied a method for flagging differential item functioning (DIF) based on loss functions. Builds on earlier research that led to the development of an empirical Bayes enhancement to the Mantel-Haenszel DIF analysis. Tested the method through simulation and found its performance better than some commonly used DIF classification systems. (SLD)

Descriptors: Bayesian Statistics, Identification, Item Bias, Simulation, Test Items

Publication Type: Journal Articles; Reports - Evaluative

Education Level: N/A

Audience: N/A

Language: English

Sponsor: N/A

Authoring Institution: N/A

Grant or Contract Numbers: N/A

Author Affiliations: N/A