Systematic equating error with the randomly-equivalent groups design: An examination of the equal ability distribution assumption
2008 (English)Report (Other academic)
In this study the equal ability distribution assumption associated with therandomly-equivalent groups equating design was investigated in the contextof a selection test for admission to higher education. Test-takers’ scores onanchor items from two subtests were estimated using information abouttest-taker performance on the regular subtests. The results showed that theestimated anchor test scores varied sufficiently so that the equal ability distributionassumption could be questioned. Consequently, we call for morecaution when applying the randomly-equivalent groups design in the equatingof tests. Equal ability groups is a convenient assumption to make but itcan lead to systematic bias in the equating of test scores and this study providesa demonstration of that point.
Place, publisher, year, edition, pages
Umeå: Institutionen för beteendevetenskapliga mätningar , 2008. , 20 p.
, EM, ISSN 1103-2685 ; 61
equating error, randomly-equivalent groups design, anchor tests, college admission tests
IdentifiersURN: urn:nbn:se:umu:diva-20426OAI: oai:DiVA.org:umu-20426DiVA: diva2:208680
Distributor:Beteendevetenskapliga mätningar, 90187, Umeå