Systematic equating error with the randomly-equivalent groups design: An examination of the equal ability distribution assumption
(English)Manuscript (preprint) (Other academic)
The equal ability distribution assumption associated with the randomly-equivalent groups equating design was investigated in the context of a selection test for admission to higher education. Test-takers’ scores on anchor items from two subtests were estimated using information about test-taker performance on the regular subtests. The results indicated that the anchor test item performance varied sufficiently so that the equal ability distribution assumption could be questioned. Consequently, our conclusion is that more caution when applying the randomly-equivalent groups design in the equating of tests is needed. Assuming equal ability groups is a convenient assumption to make but it also can lead to systematic bias in the equating of test scores and this study provides a demonstration of that point.
equating error, randomly-equivalent groups design, anchor tests, college admission tests
IdentifiersURN: urn:nbn:se:umu:diva-25430OAI: oai:DiVA.org:umu-25430DiVA: diva2:231762