ON THE USE OF THE GENERALIZED-T AND GENERALIZED RANK-SUM STATISTICS IN MEDICAL-RESEARCH

We have used Monte Carlo methods to compare the type I error propertie s of the conditional and unconditional versions of the generalized t a nd the generalized rank-sum tests to those of the independent samples t and Wilcoxon rank-sum tests. Results showed inflated type I errors f or the conditional generalized tests but not for the unconditional tes ts. We also compared the power of the unconditional generalized tests to that of the t and Wilcoxon tests under a variety of conditions. Res ults showed the generalized tests to be much more efficient than their traditional counterparts in some circumstances, but substantially les s powerful in others. Based on these and other considerations, we conc lude that the application of these newer statistics in medical researc h needs further consideration.