Nonparametric methods: The Wilcoxon Signed-Rank Test

In the last exercise, you performed a sign test. However sometimes you will have ordinal data where people have been rating two different products and where the difference in ratings matters. Let's assume we work for a consumer goods company and we have been releasing two different beers. Our first beer is an IPA and our second is a wheat beer. We allow our test panel of customers to rate both beers on a scale from 1 - 5. In order to analyze this data, we can use A Wilcoxon signed rank test. This is the equivalent of a one-sample t-test.

Before we move on to doing a Wilcoxon signed-rank test in R, let's first get our hypotheses right. To do so, let's work with our example of rating two different types of beer: the IPA and the wheat beer. In this example we have preliminary evidence that consumers prefer our IPA to our wheat beer. To gain further evidence for this, we are presenting both beers to our test panel. Choose the most appropriate set of hypotheses using this example given the fact that we want to indicate directionality in our hypotheses.

Possible Answers

H0: The beers are equally appreciated by our test panel. H1: Our IPA receives higher appreciation from our test panel than the wheat beer.
H0: The beers are equally appreciated by our test panel. H1: Our beers are not equally appreciated by our test panel.
H0: The beers are not equally appreciated by our test panel. H1: The beers are equally appreciated by our test panel.

Exercise

Nonparametric methods: The Wilcoxon Signed-Rank Test

Instructions

Possible Answers