Some Methods and Evaluation for Linking and Equating with Small Samples

Abstract

ABSTRACT The purpose of the current article is to introduce the equating and evaluation methods used in this special issue. Although a comprehensive review of all existing models and methodologies would be impractical given the format, a brief introduction to some of the more popular models will be provided. A brief discussion of the conditions required for equating precedes the discussion of the equating methods themselves. The procedures in this review include the Tucker method, mean equating, nominal weights mean, simplified circle arc, identity equating, and IRT/Rasch model equating. Models shown that help to evaluate the success of the equating process are the standard error of equating, bias, and root-mean-square error. This should provide readers with a basic framework and enough background information to follow the studies found in this issue.