Empirical Selection of Anchors for Tests of Differential Item Functioning

Autor/in	Woods, Carol M.
Titel	Empirical Selection of Anchors for Tests of Differential Item Functioning
Quelle	In: Applied Psychological Measurement, 33 (2009) 1, S.42-57 (16 Seiten) PDF als Volltext Verfügbarkeit
Sprache	englisch
Dokumenttyp	gedruckt; online; Zeitschriftenaufsatz
ISSN	0146-6216
DOI	10.1177/0146621607314044
Schlagwörter	Test Results; Testing; Item Response Theory; Test Bias; Statistical Bias; Maximum Likelihood Statistics; Error of Measurement; Simulation; Multiple Regression Analysis; Evaluation Methods + Suchen Sie Ihr Suchwort? Testdurchführung; Testen; Item-Response-Theorie; Testkritik; Messfehler; Simulation program; Simulationsprogramm
Abstract	Differential item functioning (DIF) occurs when items on a test or questionnaire have different measurement properties for one group of people versus another, irrespective of group-mean differences on the construct. Methods for testing DIF require matching members of different groups on an estimate of the construct. Preferably, the estimate is based on a subset of group-invariant items called designated anchors. In this research, a quick and easy strategy for empirically selecting designated anchors is proposed and evaluated in simulations. Although the proposed rank-based approach is applicable to any method for DIF testing, this article focuses on likelihood-ratio (LR) comparisons between nested two-group item response models. The rank-based strategy frequently identified a group-invariant designated anchor set that produced more accurate LR test results than those using all other items as anchors. Group-invariant anchors were more difficult to identify as the percentage of differentially functioning items increased. Advice for practitioners is offered. (Contains 4 tables.) (As Provided).
Anmerkungen	SAGE Publications. 2455 Teller Road, Thousand Oaks, CA 91320. Tel: 800-818-7243; Tel: 805-499-9774; Fax: 800-583-2665; e-mail: journals@sagepub.com; Web site: http://sagepub.com
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2017/4/10