Experiments were designed to determine whether genetic predisposition to diabetes or overt diabetes in the identical genotype had an effect on the level of neutralization antibodies to coxsackievirus B4 (CB4). Quantitation of neutralization antibody (NT) levels against CB4 was performed using both the 50% endpoint procedure and the multivariate Wilcoxon rank sum test. The results of these experiments demonstrate that the use of the multivariate procedure for the analysis of neutralization antibody levels provides both quantitative and qualitative information not evident when only the classical 50% end point procedure is used. Moreover, when comparison on antibody levels between different groups is being considered, the power of the comparison is markedly greater using the multivariate Wilcoxon rank sum test results. The present report provide an illustration of the difference in the qualitative and quantitative information obtained by the end-point procedure and the more comprehensive multivariate procedure for the analysis of neutralization antibody levels in diabetic mutant mice infected with coxsackievirus B4.