The multivariate multilevel model (MVMM) is an extension of the univariate multilevel model (MLM) that may be used in the presence of multiple outcomes. In a two-group cluster randomized design, one approach is to apply the MLM separately to each outcome variable whereas an alternate approach is to use the MVMM, which incorporates all outcomes simultaneously in a single analysis model. This Monte Carlo study investigated the degree to which results from the two models differ across a set of conditions that can be considered to favor the use of univariate analysis. Our results showed there were no differences in the performance of the MLM and MVMM with respect to estimation bias, power, and Type I error rate. We discuss the implications of these findings for applied researchers.