This study provides a direct comparison of the ANOVA F-test and the Kruskal-Wallis Test, a commonly used nonparametric alternative, in terms of type I error rate and power performance. The Monte Carlo simulation expands upon previous studies by varying all common relevant factors simultaneously. The study included normal and nonnormal distributions, homogeneous and heterogeneous variance structures, balanced and unbalanced sample size designs, and also four specific noncentrality structures. Results indicate that the Kruskal-Wallis test is optimal to its parametric counterpart in many specific types of configurations.