Estimating IRT Equating Coefficients for Tests with Polytomously and Dichotomously Scored Items.

Abstract

Recent years have seen growing use of tests with mixed item formats, e.g., partly containing dichotomously scored items and partly consisting of polytomously scored items. A matching two test characteristic curves method (CCM) for placing these mixed format items on the same metric is described and evaluated in this paper under a common-item linking design in which tests containing a set of common items are administered to two groups of examinees. The BIAS statistics of the recovery equating coefficients across all research conditions are usually expected to be close to zero, as happened in this simulation, and the corresponding root mean squared error (RMSE) statistics, in general, are relatively small. The shapes of the empirical sampling distributions of the estimated equating coefficients obtained from a variety of conditions are symmetric-bell shapes with small variances and few or no outliers or other abnormalities. The CCM is capable of producing accurate transformation of parameter estimates when applied to a test with mixed items as well as to only dichotomously scored or polytomously scored tests. (Contains 3 tables, 4 figures, and 35 references.) (Author/SLD) ******************************************************************************** Reproductions supplied by EDRS are the best that can be made from the original document. ******************************************************************************** Estimating IRT Equating Coefficients for Tests with Polytomously and Dichotomously Scored Items