To assess the effect on discriminant analysis in terms of correct classification into two groups, the following parameters were systematically altered using Monte Carlo techniques: sample sizes; proportions of one group to the other; number of independent variables; and covariance matrices. The pairing of the off diagonals (or covariances) with the different numbers of variables and several different pairs of sample sizes within differing proportions required 396 Monte Carlo studies, each using 100 simulations. Results suggest that the proportion of one sample size to another is a major influence on correct classification. Correct classification was influenced by neither heterogeneity of covariance matrices, the number of variables used, nor a specific sample size when the proportional relationship between two samples was constant. The greatest hit rate occurred in the group with the largest sample. Statistics also found to be robust to heterogeneity of covariance matrices included: (1) eigenvalues; (2) canonical correlations; (3) Pillai's trace; and (4) Wilks' Lambda. Researchers using discriminant analysis to classify will be able to take into account and use the robustness of these statistics. Six bar graphs illustrate the study findings. (SLD)