Equating Between a Latent and a Rating Score Scales Under The Common Examinees Design

Abstract

In this research, we proposed an equating method under the common examinees design in case one test is on a latent scale and another test is on an original rating score scale. This method is useful when the number of examinees of either test is not sufficient to apply latent trait model such as item response theory. This situation above is not new in Japan, for example, national center test for university admissions, and so on. In addition, the accuracy of this method was examined by simulation studies, and we showed the procedure for creating a score conversion table through real data analysis.