One-Way ANOVA with within-Subjects Data: The Case of Non-Random Missingness in Skewed Distributions

Monte Carlo study examining the Type-I error rates for the one-way and repeated measures ANOVA under different conditions. Results found do not indicate even a moderate advantage for the one-way ANOVA in terms of statistical power. In fact, across a majority of conditions, the repeated-measures ANOVA exhibited superior performance.