ERIC Number: ED241540
Record Type: RIE
Publication Date: 1982-Mar
Reference Count: 0
Test Equating: Relevant Issues and a Review of Recent Research.
Skaggs, Gary; Lissitz, Robert W.
Equating studies using item response theory (IRT) are reviewed. The most well-known papers, as well as a sampling of lesser-known studies, are included. Accompanying tables list the papers and classify them according to the test used, models used, test length and type, sample size and type, method of assessment, equating design, and kinds of comparisons made. A majority of the equating research has focused on the Rasch, or one parameter logistic, model. Initial Studies using the Rasch model investigated the invariance properties of the model: person-free item calibration and item-free person measurement. With tests of similar difficulty and samples of comparable ability, the research suggests that Rasch horizontal equating provides reasonable results. Research on other IRT models has focused on comparing different strategies using the same data set. With regard to vertical equating, most of the research has demonstrated the superiority of the three-parameter model over the Rasch model. Additional equating studies using Monte Carlo methods are reviewed. Finally, four issues relevant to test equating are discussed: assessing the adequacy of equating, sources of equating error, multidimensionality, and out-of-level testing. (PN)
Publication Type: Speeches/Meeting Papers; Information Analyses
Education Level: N/A
Authoring Institution: N/A
Identifiers: One Parameter Model; Rasch Model; Three Parameter Model
Note: Paper presented at the Annual Meeting of the American Educational Research Association (65th, Los Angeles, CA, April 13-17, 1981).