A Matter of Test Bias in Educational Policy Research: Bringing the Context into Picture by Investigating Sociological/Community Moderated (or Mediated) Test and Item Bias

Autor/inn/en	Zumbo, Bruno D.; Gelin, Michaela N.
Titel	A Matter of Test Bias in Educational Policy Research: Bringing the Context into Picture by Investigating Sociological/Community Moderated (or Mediated) Test and Item Bias
Quelle	In: Journal of Educational Research & Policy Studies, 5 (2005) 1, S.1-23 (23 Seiten) PDF als Volltext kostenfreie Datei Verfügbarkeit
Sprache	englisch
Dokumenttyp	gedruckt; online; Zeitschriftenaufsatz
ISSN	1934-6875
Schlagwörter	Test Bias; Multiple Choice Tests; Mathematics Tests; Numeracy; Foreign Countries; Measurement; Grade 4; Grade 7; Grade 10; Context Effect; Gender Differences; Standardized Tests; Rural Urban Differences; Family Income; Case Studies; Canada + Suchen Sie Ihr Suchwort? Testkritik; Multiple choice examinations; Multiple-choice tests, Multiple-choice examinations; Multiple-Choice-Verfahren; Rechenkompetenz; Ausland; Messverfahren; School year 04; 4. Schuljahr; Schuljahr 04; School year 07; 7. Schuljahr; Schuljahr 07; Geschlechterkonflikt; Standadised tests; Standardisierter Test; Stadt-Land-Beziehung; Familieneinkommen; Case study; Fallstudie; Case Study; Kanada
Abstract	The present study introduces and demonstrates a new methodology for item and test bias studies: "moderated differential item functioning (DIF)." This technique expands the DIF methodology to incorporate contextual and sociological variables as moderating or mediating effects of the DIF. Specifically, this paper explores differential domain functioning (DDF)--the focus of interpretation for this test is on the "domain" rather than the item. This moderated DDF effect is demonstrated on a multiple-choice and constructed-response provincial assessment test that was designed to match a specified mathematics curriculum. Participants were 45,728 grade four students, 45,022 grade seven students, and 43,525 grade 10 students in British Columbia, Canada. The data from these participants was narrowed down to create four contrast groups of communities that reflect differences in contextual variables: rural low-income, rural affluent, urban low-income, and urban affluent. Gender DDF was explored using a general linear statistical model. After statistically matching males and females on their mathematical ability, gender DDF was moderated by the contextual variables. Thus, this "moderation" approach allows one to investigate the effect of sociological, community-based contextual variables that may help one understand the complex functioning of DIF in large scale testing. In other words, what the authors are advocating is to take a more "sociological" and "ecological" approach to help educators understand differences in item and test performance. (Contains 5 tables and 1 figure.) (As Provided).
Anmerkungen	National Office for Research on Measurement and Evaluation Systems. University of Arkansas, 346 North West Avenue, 302 WAAX, Fayetteville, AR 72701. Tel: 479-575-5593; Fax: 479-575-5185; e-mail: normes@uark.edu; Web site: http://normes.uark.edu/erps/resources.html
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2017/4/10