Evaluating the effects of multidimensionality on IRT true-score equating

The performance of the item response theory (IRT) true-score equating metho d is examined under conditions of test multidimensionality. It is argued th at a primary concern in applying unidimensional equating methods when multi dimensionality is present is the potential decrease in equity (Lord, 1980) attributable to the fact that examinees of different ability are expected t o obtain the same test scores. In contrast to equating studies based on rea l test data, the use of simulation in equating research not only permits as sessment of these effects but also enables investigation of hypothetical eq uating conditions in which multidimensionality can be suspected to be espec ially problematic for test equating. In this article, I investigate whether the IRT true-score equating method, which explicitly assumes the item resp onse matrix is unidimensional, is more adversely affected by the presence o f multidimensionality than 2 conventional equating methods-linear and equip ercentile equating-using several recently proposed equity-based criteria(Th omasson, 1993). Results from 2 simulation studies suggest that the IRT meth od performs at least as well as the conventional methods when the correlati on between dimensions is high (greater than or equal to 0.7) and may be onl y slightly inferior to the equipercentile method when the correlation is mo derate to low (less than or equal to 0.5).