Mann-Whitney Non Parametric U Test
The preferred non-parametric method for unpaired samples is the Mann-Whitney non parametric hypothesis test or Mann-Whitney test (it is also called as Wilcoxon Rank Sum Test or the Mann-Whitney Wilcoxon Test) and thus the non parametric solution to evaluating two independent datasets comparable to the Student’s T-test.
The Mann-Whitney U test is a non-parametric statistical test used to compare two independent samples. It assesses whether the distribution of values in one sample is stochastically greater or less than the distribution of values in another sample. This test does not specifically compare sample means, but rather it evaluates whether the two samples come from the same population or if one tends to have higher values than the other.
How Decision Rules Differ Between Mann-Whitney U and Parametric t-Tests
It’s important to understand how the Mann-Whitney U test differs in decision-making from parametric tests like the independent or paired t-test. While all of these tests aim to determine if two groups differ, their approaches and interpretations are different:
- Parametric t-Tests: These tests compare means and assume that the underlying data follows a normal distribution. A small p-value in a t-test typically leads to the conclusion that there is a statistically significant difference in average values between the two groups.
- Mann-Whitney U Test: This test does not compare means. Instead, it ranks the data and tests whether values in one group tend to be higher or lower than those in the other group. It makes no assumption about the shape of the distribution, making it a safer option when data is skewed or ordinal. A significant p-value here suggests that one population tends to yield higher (or lower) values than the other—but not necessarily that their medians or means differ.
So, while both tests help you reject or accept a null hypothesis, the t-test focuses on differences in central tendency (mean), and the Mann-Whitney focuses on differences in rank distributions.
A few researchers also interpret the Mann-Whitney test will help to compare the medians of the two populations. Usually, parametric tests will compare the means (Null hypothesis: μ1=μ2) between the independent groups. Whereas, the null hypotheses and the two-sided hypotheses non-parametric tests can be indicated as follows:
- Null Hypothesis (H₀): The two populations come from the same distribution.
- Alternative Hypothesis (H₁): The two populations come from different distributions.
Note: Unlike t-tests, Mann-Whitney does not assume a difference in means, only a difference in distribution or tendency.
The Mann-Whitney test is regularly performed as a two-sided test, therefore the investigative hypothesis indicates that the two populations are not equal, instead of specifying the directionality.
Assumptions of the Mann-Whitney:
- The sample drawn from the population is random.
- Independence within the samples and also mutual independence is assumed. That means that an observation is in one group or in the other (it cannot be in both).
- Identical (non-normal) distributions
- An Ordinal measurement scale is assumed.
Uses of Mann-Whitney
The Mann-Whitney U test can be used in many industries, but it is most frequently applied in healthcare, nursing, business, and related fields. In medicine, it is a more efficient method to know the effect of two medicines and whether they are equal or not. It is also used to know whether or not a particular medicine cures an ailment or not.
Procedure to conduct Mann-Whitney test
- Combine the two data samples into a single dataset and sort them in ascending order.
- Assign ranks to all observations.
- Separate the ranks back into their original groups.
- Calculate the U statistic using the sum of ranks for each group.
- Find the critical U value from the Mann-Whitney table based on your sample sizes and significance level.
- Compare the calculated U value to the critical value and draw a conclusion.
Calculation of Mann-Whitney Test
Case 1: For comparing two small sets of observations (samples sizes less than 20)
U1 = n1n2+0.5n1(n1+1)-R1
U2 = n1n2+0.5n2(n2+1)-R2
Where U1+U2=n1n2
- n1 is the observations from the first population
- n2 is the observations from the second population
- R1 is the sum of observation ranks for the first population
- R2 is the sum of observation ranks for the second population
Finally, calculate the U statistic as the smaller U1 and U2. U= min (U1, U2)
Case2: Normal approximation and tie correction
For sample sizes above ~20, the distribution of U rapidly approaches the normal distribution
U mean = μU =0.5 n1n2

Example of Mann-Whitney Test
A researcher, while conducting studies on the Biomass of various trees, wished to determine if there was a difference in the biomass of male and female Juniper trees. So, he randomly selected 6 individuals of each gender from the field. He dries them to constant moisture, chips them, and then weighs them to the nearest kg.

- H0: There is no difference between the biomass of male and female Juniper trees– Biomassmale = Biomassfemale (medians are equal)
- H1: There is a difference between the biomass of male and female Juniper trees– Biomassmale ≠ Biomassfemale (medians are not equal)
The data are ratings (ordinal data), and hence a non-parametric test is appropriate – the Mann-Whitney U test (the non-parametric counterpart of an independent measures t-test).

- n1 =6
- n2 =6
- R1 = 1.5 + 3 + 5.5 + 7.5 + 4 + 1.5 = 23
- R2 = 5.5 + 11 + 10 + 12 + 9 + 7.5 = 55
And then calculate U1 and U2
- U1 = n1n2+0.5n1(n1+1)-R1 = 6 *6 + 0.5 * 6(7) – 23 = 34
- U2 = n1n2+0.5n2(n2+1)-R2 = 6 * 6 + 0.5 * 6(7) – 55 = 2
U= min (U1, U2) = 2
Mann-Whitney Table: For Two-Tailed Test at a 5% Significance Level

Ucritical = 5
Calculated U is a value less than the critical value of U for a 0.05 significance level. Ucalculated < Ucritical . Hence, we can reject the null hypothesis.
Therefore, we reject the null hypothesis and conclude that there is a statistically significant difference in the biomass distributions between male and female Juniper trees.
Six Sigma Black Belt Certification Mann-Whitney Test Questions:
Question: Which of the following test is calculated by ranking all the participant’s scores from lowest to highest and adding up the ranks separately for each of the groups?
(A) Wilcoxon signed rank test
(B) Kruskal Wallis test
(C) Mann-Whitney test
(D) Friedman Test
Answer:
When you’re ready, there are a few ways I can help:
First, join 30,000+ other Six Sigma professionals by subscribing to my email newsletter. A short read every Monday to start your work week off correctly. Always free.
—
If you’re looking to pass your Six Sigma Green Belt or Black Belt exams, I’d recommend starting with my affordable study guide:
1)→ 🟢Pass Your Six Sigma Green Belt
2)→ ⚫Pass Your Six Sigma Black Belt
You’ve spent so much effort learning Lean Six Sigma. Why leave passing your certification exam up to chance? This comprehensive study guide offers 1,000+ exam-like questions for Green Belts (2,000+ for Black Belts) with full answer walkthroughs, access to instructors, detailed study material, and more.
Comments (15)
“Mann-Whitney test is a non-parametric test that is to compare two sample means”
How come Non-parametric test and used to compare means?
Hi Mohamed,
I think you’re asking why you would use a non-parametric test to test means. I’ll answer that. If you had a different question in mind, just leave another comment.
You would use a non-parametric test of means when the distribution you’re testing against isn’t normal.
Best, Ted.
“Calculated U is value less than the critical value of U for a 0.05 significance level. Ucalculated Shouldn’t it means reject null instead?
I agree with you. The following sentence means that the null hypothesis was rejected.
> So, we can say that seem like there is a highly significant difference between biomass of male and female Juniper trees.
Thank you.
Those disagreements between the conclusions in the same articles are really annoying. Led to massive misunderstanding!
Thank you, we have corrected the statistical conclusion.
Hi Ted
I have juste a question please
I observe that there is a lot of reference of certification
Like IASSC, ASQ, 6 sigma study , councel six sigma
in your point of views which certifications better ,
and what reference should we avoid
Youssef, It really depends on the individual. I have a comprehensive article here.
I am somewhat confused.
In Paired T Test, when the calculated statistic is below the Critical value, we fail to reject it.
In non-parametric Mann Whitney, when calculated statistic is below the Critical value, we reject it?
Thank you for your thoughtful question, Ashwin. The confusion you’re experiencing is quite common, as the decision rules for parametric and non-parametric tests can differ based on how their test statistics are defined and how critical values are interpreted.
For the Paired T-Test, the test statistic is compared to a critical value from the t-distribution. If the absolute value of your calculated t-statistic is greater than the critical value, you reject the null hypothesis; if it is less, you fail to reject it.
With the Mann-Whitney U test, the logic is a bit different, and it depends on how the U statistic and critical value are defined in your tables or software. Typically, if your calculated U statistic is less than or equal to the critical value, you reject the null hypothesis. This is because a smaller U suggests a greater difference between groups than would be expected by chance.
So, your understanding is correct: for the Mann-Whitney test, a lower U statistic (below the critical value) usually means you reject the null hypothesis. For the Paired T-Test, a lower t-statistic (in absolute value) means you fail to reject the null hypothesis. The key is to always check the specific decision rule for the test you’re using, as the direction of comparison can vary.
If you’d like to deepen your understanding of hypothesis testing and the differences between these tests, you might find our resources on the Green Belt Course and Black Belt Course helpful.
Warm regards,
Ted
SixSigmaStudyGuide.com
aren’t we supposed to compare the medians here?
This test has three types: the exact test, the median confidence interval and the advanced one.
Hello mohamed ramzi bentoumi,
The Mann-Whitney U test is used to compare differences between two independent groups. It is the non-parametric counter part to the t-test for the independent samples. t-test checks is there a difference in mean. In other hand, Mann-Whitney U test checks is the difference in the rank sum. When we have two groups, we check is there a difference in rank sum of the first group and the second group. Please refer article on how to calculate the rank sum.
Thanks
which test of non parametric are independent and which are dependent in nature
Hi meshach
Independent non-parametric tests:
Mann-Whitney U test
Kruskal-Wallis test
Friedman test
Kolmogorov-Smirnov test
Dependent non-parametric tests:
Wilcoxon signed-rank test
Sign test
McNemar’s test
Friedman’s test with repeated measures
Thanks