Univariable Analysis of an Ordinal Dependent Variable

There are some types of information in health research that are naturally measured on an ordinal scale. Statistical methods for ordinal dependent variables can be used to analyze data that naturally occur on an ordinal scale, but this is not the most common use of these methods. Nonparametric methods are an alternative to statistical methods designed for continuous data. This chapter discusses the Wilcoxon signed-rank test. The Wilcoxon signed-rank test is the nonparametric parallel to the paired t-test. Continuous data can be converted to an ordinal scale to circumvent some of the assumptions required for the analysis of continuous dependent variables. The loss of statistical power due to use of a nonparametric test will be small and will not often affect the conclusion we draw in statistical hypothesis testing.