Nonparametric methods: The Wilcoxon rank-sum test

The Wilcoxon rank-sum test is the nonparametric equivalent of the two sample t test. Your confusion may stem from the fact that this name is really similar to the Wilcoxon signed-rank test. You can use the Wilxocon rank-sum test when you deal with two different groups and want to test whether they differ significantly. Let's take our previous example with the IPA and wheat beer. One half of our test panel will consume the IPA. The other half of the test panel will drink the wheat beer. After, each group member will report their levels of satisfaction with the drink.

Before we move on to doing a Wilcoxon rank-sum test in R, let's first get our hypotheses for this new experiment. In this example we still have preliminary evidence that the IPA is associated with thigher levels of satisfaction than our wheat beer. Choose the most appropriate set of hypotheses using this example given the fact that we want to indicate directionality in our hypotheses.