a multiple comparison procedure that is used to test for significant differences between all possible pairs of mean values on a variable for groups of research participants; it is generally applied after an
analysis of variance has determined that there is a significant difference among three or more means. The procedure involves simultaneously comparing all possible pairs of means based on a single quantity, called the
honestly significance difference (HSD), such that if the difference between any two group means exceeds the HSD, the corresponding population means are said to be significantly different from each other. Tukey’s HSD test preserves the
family-wise error rate by adjusting the overall
significance level to take into account the fact that multiple
t tests are being conducted. Also called
Tukey’s range test. [John Wilder Tukey]