Skip to main content
Two Means - Unknown, Unequal Variance
10. Hypothesis Testing for Two Samples / Two Means - Unknown, Unequal Variance / Problem 2
Problem 2

The Mann-Whitney UU test is a nonparametric test used to compare two independent samples. It is equivalent to the Wilcoxon rank-sum test and leads to the same conclusions when applied appropriately. The test statistic is calculated as:
z=Un1n22n1n2(n1+n2+1)12z=\frac{U-\frac{n_1n_2}{2}}{\sqrt{\frac{n_1n_2\left(n_1+n_2+1\right)}{12}}}
where:
U=n1n2+n1(n1+1)2RU=n_1n_2+\frac{n_1\left(n_1+1\right)}{2}-R
n1=n_1= size of Sample 11
n2=n_2= size of Sample 22
R=R=sum of the ranks for Sample 11
Use the following pulse rate data (in units of beats per minute) for two groups of athletes to calculate the zz-test statistic for the Mann-Whitney UU test.
Sample 11 (athletes) : 72,78,80,85,9072, 78, 80, 85, 90
Sample 22 (long-distance runners): 60,65,68,70,7560, 65, 68, 70, 75