When an ANOVA comparing the means of 4 groups indicates that at least one group is different from the others, a common follow-up analysis to determine which group(s) is (are) different is pairwise two-sample t-tests each assessed using i) a pooled standard deviation based on the ANOVA MSE when calculating the standard error for the differences in means and ii) a Bonferroni-corrected alpha level of 0.0083 to control the type I error rate for the overall inference to 5% .