Continuous variable analyses: Student's t-test, rank-sum test, and signed-rank test

Continuous variable analyses include a variety of potential statistical tests that can be employed depending on the relevant underlying assumptions. Parametric tests make assumptions about the underlying data distribution while nonparametric tests do not. The Student's t-test, the Wilcoxson rank-sum test (Mann-Whitney U), and the Wilcoxson signed-rank test are three of the most common statistical tests employed for continuous variable analyses. The Student's t-test is a parametric test that assumes independence of samples, random sampling, equal variances, and a normal distribution of the samples. The nonparametric Wilcoxson rank-sum test assumes both samples have the same distribution but does not require a normal distribution. For paired data or repeated measures, the parametric paired t-test or nonparametric Wilcoxson signed-rank test should be considered. Understanding the basic assumptions of each test and how to assess data samples can help clinical researchers make better inferences about populations of interest.