A randomization test is valid for any kind of sample, no matter how the sample is selected. This is an extremely important property because the use of non-random samples is common in experimentation, and parametric statistical tables (e.g., t and F tables) are not valid for such samples.

What are randomisation tests?

Randomization tests can be thought of as another way to examine data, and do not restrictive assumptions about populations. ... So let's set out by taking all of our data, tossing it in the air, and letting half of it fall in one group and the other half in the other group.

Where does the sampling distribution come from when you use a randomization test?

The randomization distribution is the histogram of all values for the statistic from all possible ways the experimental units could have been randomly assigned to groups. In the sampling model, the reason there is variability in a sample statistic is because we induced variability by taking a random sample.