Abstract

In order to compare scores from two or more related tests it is necessary for them to be reported on a common scale. Various linking methods have been developed for this purpose, depending on the statistical properties of the tests and the data collection design (cf., Dorans, Pommerich, & Holland, 2007; Kolen & Brennan, 2004); however, most of these methods are premised on the assumption that a single construct is measured within and between tests. When a multidimensional model is applied to tests measuring more than one dimension it is important to consider the comparability of dimension-specific scores. The goal of this chapter is to provide a foundation for understanding multidimensional test score linking and various issues that should be considered in this process.