Title: Estimating a unidimensional latent ability across multiple test forms with confounding difficulty and length within sub-content areas
Authors: Ki Lynn Matlock; Ronna C. Turner; Jam Khojasteh
Addresses: College of Education, Oklahoma State University, 313 Willard Hall, Stillwater, OK 74078, USA ' College of Education and Health Professions, University of Arkansas, Graduate Education Building 250, Fayetteville, AR 72701, USA ' College of Education, Oklahoma State University, 2444A Main Hall, Tulsa, OK 74106, USA
Abstract: This study investigated the effects of estimating unidimensional latent abilities for subgroups of a population across multiple test forms with confounding difficulty and number of items within sub-content areas. Examinees were grouped based on their true two-dimensional ability; estimates within subgroups were compared across test forms having equal length and average item difficulty overall, but differing numbers of items and/or difficulty within subsets. Examinees with differing true abilities across two dimensions had significantly different estimated scores across forms, depending on the alignment of true ability in each dimension with the number of items and average item difficulty of items within that dimension. These effects tended to decrease as the correlation between dimensions increased. The results of this study bring awareness to test developers for the need of controlling item specifications within sub-content areas.
Keywords: item response theory; IRT; multidimensional data; unidimensionality; test development; latent traits; multiple test forms; confounding difficulty; length; sub-content areas; item specification control.
International Journal of Quantitative Research in Education, 2016 Vol.3 No.3, pp.154 - 178
Received: 02 Apr 2015
Accepted: 24 Aug 2015
Published online: 15 Jul 2016 *