This study investigated seven methods for analyzing multivariate group differences. Bonferroni t statistics, multivariate analysis of variance (MANOVA) followed by analysis of variance (ANOVA), and five other methods were studied using Monte Carlo methods. Methods were compared with respect to (1) experimentwise error rate; (2) power; (3) number of Type 1 errors in experiments with at least one error; and (4) for experiments with at least one false univariate hypothesis, the probability of rejecting at least one of the true hypotheses. One method emerged as having the best all around performance. This method used repeated T-squared statistics and removed the variable with maximum significant F statistic, providing a good balance between power and Type 1 errors. It consisted of the following steps: (1) MANOVA on p variables followed by ANOVAs; (2) reject the hypothesis for the variable with the largest significant F statistic and remove that variable; (3) MANOVA on p-1 variables; (4) repeat Step 2 with p-1 variables; (5) MANOVA on p-2 variables...and so on until no MANOVAs are significant, no ANOVAs are significant, or there are no variables left. (Author/PN)