The determination of the U-statistic in the Mann-Whitney U test is the smaller of the two values of group A and group B as calculated by the following:
U-statistic Group A = n1n2 + {[n1(n1 + 1)]/2} - R1
= (9)(10) + {[9(9 + 1)]/2} - 97.5
= 90 + 45 - 97.5
= 37.5
U-statistic Group A = n1n2 + {[n2(n2 + 1)]/2} - R2
= (9)(10) + {[10(10 + 1)]/2} - 92.5
= 90 + 55 - 92.5
= 52.5
Note: n1 = sample size of group A; n2 = sample size of group B; R1 = summed rank score of group A; R2 = summed rank score of Group B.
The smaller of the two values calculated is 37.5. The next step is to compare the observed value of U against the critical value of U. The observed value of U is statistically different if it is less than or equal to the critical U.
I'll let you take it from here to draw your final conclusion.
for some independent data using the mann whitney test
group a has rank sum of 97.5 with group size = 9
group b has a rank sum of 92.5 with a group size of 10
from the data table i have the two critical values for p=0.05 are 65 and 115
i know im supposed to compare them and decide whether there is a significant difference between them but im not sure which way round they go could you explain it please
thank you
2 answers
One correction: the second calculation should read as U-statistic Group B.