Local Equating of Cognitively Diagnostic Modeled Observed Scores

Abstract

A local equating method of cognitively diagnostic modeled observed score was introduced to address the problem of equating without anchor test as well as the satisfaction of equating criteria. The criterion of test construct was redefined on attribute level. Then, a true family of equating transformations given attribute mastery pattern was derived from the criterion of equity, which is the main concern of local equating. The results of simulation studies showed a substantial bias for the traditional equipercentile equating under the equivalent groups design but much smaller bias for the local method. The proposed method also showed satisfactory performances under the nonequivalent groups design even without anchor test. The local method appeared to be quite robust with respect to changes in the sample size, but its biases became larger with increasing model misfit.

Estimating an Observed Score Distribution From a Cognitive Diagnostic Model

“…Still, these results do show some promise for the proposed estimation method. This method of score distribution estimation has been used in at least two studies so far: by Skaggs, Hein, and Wilkins (in press) for standard setting and by Xin and Zhang (2015) for test equating. However, it must be pointed out that research on CDMs is relatively new, and there are as yet few actual applications of CDMs in large-scale testing.…”

An Item‐Level Expected Classification Accuracy and Its Applications in Cognitive Diagnostic Assessment

“…Fortunately, Xu and von Davier () proposed three linking strategies by fixing the common items parameters or setting the common items to be equal across test forms under a general diagnostic model containing discrete latent skills. For assumptions of the common linking, Xin and Zhang () suggested that two test forms to be equated should involve the same set of attributes with identical hierarchy.…”