The mean rank is the average of the ranks for all observations within each sample. Minitab uses the mean rank to calculate the H-value, which is the test statistic for the Kruskal-Wallis test.
To calculate the mean rank, Minitab ranks the combined samples. Minitab assigns the smallest observation a rank of 1, the second smallest observation a rank of 2, and so on. If two or more observations are tied, Minitab assigns the average rank to each tied observation. Minitab calculates the mean rank for each sample.
When a group's mean rank is higher than the overall average rank, the observation values in that group tend to be higher than those of the other groups.