When various forms of a test are administered to different groups of individuals, the scores obtained are not directly comparable, due to differences in the difficulty of the items. To adjust for such differences, several equating methods have been proposed in the literature. Recently, a new likelihood-based method was developed to equate a large number of test forms. Following an Item Response Theory (IRT) approach, the proposal differs from the previous ones as it accounts for the heteroskedasticity and the correlation of the item parameter estimates. In this paper, we explore the possibility of treating the item parameter estimates as uncorrelated through an extensive simulation study. The results indicate a negligible loss in estimator efficiency, while offering a significant gain in computational time.