Test Equating Under the NEAT Design: A Necessary Condition for Anchor Items

Mroch, Suh, Kane, & Ripkey (2009); Suh, Mroch, Kane, & Ripkey (2009); and Kane, Mroch, Suh, & Ripkey (2009) provided elucidating discussions on critical properties of linear equating methods under the nonequivalent groups with anchor test (NEAT) design. In this popular equating design, two test forms are administered to different groups of examinees and an anchor subtest consisting of several items is part of the two forms. This subtest, often referred to as anchor test, comprises the same items that are used in the two groups and is typically utilized subsequently to link the resulting scores on the two forms. This note is concerned with what may well be viewed as a necessary condition for an application of the NEAT design. This condition seems not to have received the attention it deserves in much of the literature on test equating. The condition is the so-called measurement invariance (e.g., Meredith, 1993; Meredith & Horn, 2001; Muthén & Muthén, 2008) with regard to the items comprising the anchor test. Copyright © Taylor & Francis Group, LLC.