Expected Classification Accuracy

Abstract

Every time we make a classification based on a test score, we should expect some number of misclassifications. Some examinees whose true ability is within a score range will have observed scores outside of that range. A procedure for providing a classification table of true and expected scores is developed for polytomously scored items under item response theory and applied to state assessment data. A simplified procedure for estimating the table entries is also presented. Accessed 19,668 times on https://pareonline.net from August 29, 2005 to December 31, 2019. For downloads from January 1, 2020 forward, please click on the PlumX Metrics link to the right.