Constructed‐response Dif Evaluations for Mixed‐format Tests

Abstract

In this study, differential item functioning (DIF) methods utilizing 14 different matching variables were applied to assess DIF in the constructed‐response (CR) items from 6 forms of 3 mixed‐format tests. Results suggested that the methods might produce distinct patterns of DIF results for different tests and testing programs, in that the DIF methods' results might be similar for tests with multiple‐choice (MC) and CR scores that are similar in their measurement characteristics but would exhibit larger variations for tests with MC and CR scores having more distinct measurement characteristics. Impact measures of the MC and CR scores appeared to be a useful basis for indicating the scores' measurement similarity, for predicting the variations of DIF results from using these scores as matching variables, and possibly for indicating the most appropriate DIF method and matching variable for a particular test. The results are described in terms of their implications for research and practice.

A Review of Developments and Applications in Item Analysis

“…2.16 have been considered for polytomously scored items by ETS researchers, including Dorans and Schmitt (1993), Moses et al (2013), and Zwick et al (1997). At the time of this writing, there is great interest in developing more innovative items that utilize computer delivery and are more interactive in how they engage examinees.…”