A Mann-Whitney U-Test compares the means between two unrelated groups, such as comparing the difference in heights between Cats and Yorkshire Terriers.
It is considered a non-parametric test and therefore is suitable for non-parametric data. To check if your data is parametric, please check our dedicated guide: Parametric or Not Guide (PDF)
If your data is parametric you should consider using an Independent-Samples t-Test
Click Analyze > Nonparametric Tests > Legacy Dialogs > 2 Independent Samples
Within the Two-Independent-Samples box, select the test variable or dependent variable you are analysing and move it to the ‘Test Variable List’ box. Then move your independent/grouping variable into the “Grouping Variable” box.
Select Define Groups and specify the groups you are going test. In this example I am testing Cats which are labelled 1 against Yorkshire Terroirs which are labelled 2, therefore my groups are numbered 1 and 2.
Select Options and click Descriptive Statistics
Click "Continue" then "OK"
SPSS will generate three tables, for this test we need two, the Descriptive Statistics and the Test Statistics:
This table shows a selection of descriptive statistics: the sample size of each group (N), the mean of each group (Mean), and the standard deviation of each group (Std. Deviation), best practice is to report them all.
This table shows the specific test results including the Mann-Whitney U statistic (U), the Wilcoxon Statistic (W), the z-score (Z) and two different p-values / significance values. For the vast majority of applications, you will need to report the Asymp. Sig (2-tailed) version of the p-value.
Heights of Cats and Yorkshire Terriers were compared. On average the Yorkshire Terriers (Mdn = 29.27) were taller than the Cats (Mdn = 28.00) however a Mann-Whitney U test indicated that this difference was not statistically significant, U (N_{Cats}=11, N_{Terriers}=11) = 35.5, z = -1.67, p = .094.