Item response theory test equating in health sciences education

Abstract

In the context of health sciences education, and education in general, the knowledge or ability of one or several subjects in a specific area is frequently compared using different forms of a test, or by means of different instruments aimed at measuring this knowledge or ability. In such cases, test scores must be equated so that they can be properly compared. The present article aims to explain the equating of scores within the framework of item response theory (IRT), special emphasis being placed on its application in the field of health sciences education. Although there are many data collection designs that can be used within the framework of score equating, the three most widely used designs are single group, randomly equivalent groups and non-equivalent groups with anchor test. Likewise, there are different equating procedures within IRT, such as the mean/mean and mean/sigma methods, or methods based on the characteristic curve. The equating of scores must ensure that examinees taking a test do so under the same conditions; however, the procedure is also highly useful in creating a bank of items that can be used subsequently in new tests.