**Mann-Whitney Test (Non-parametric)**

**What Is It?**

A Mann-Whitney test is used when you have a small sample size and equivalent to a 2 sample t-test. It is used when you don't feel that you can assume that the data comes from a normal distribution (so especially relevant for ordinal datasets). It is used to test differences in the medians of the populations.

**How Do You Carry One Out In SPSS?**

For the purposes of this illustration we are going to assume that we are testing 2 different groups (Group 1 and Group 2) of independent data against a dependent variable.

Once you have entered your data into SPSS you need to go do

**ANALYSE > NONPARAMETRIC TESTS > LEGACY DIALOGS > 2 INDEPENDENT SAMPLES**. This will bring up the Two Independent Samples Test dialogue box which should show your 2 datasets in the left hand box. Firstly you need to check that Mann-Whitney U is ticked at the bottom of the dialogue, not any of the other tests.

Next you take your dependent variable and move it into the Test Variable List: box on the right hand side of the dialogue box, and then move your independent variable into the Grouping Variable: box which is just below the Test Variable List: box. Once that is done click Define Groups and it will ask you to define Group 1 and Group 2. Put a 1 into Group 1 and a 2 into Group 2 and click Continue.

A final dialogue box will appear presenting some other options for the test results (like would you like quartiles) which aren't Mann-Whitney related and then you can click Continue to run the test.