A clinical investigator is studying the effect of a newly developed ACE inhibitor on serum potassium levels. The study enrolls 50 adult patients with essential hypertension. Each patient has their baseline serum potassium level measured prior to starting the medication. After four weeks of daily therapy, a second serum potassium measurement is obtained for each patient. The investigator wants to evaluate if there is a statistically significant change in serum potassium levels from baseline. Depending on whether the data follows a normal distribution, which of the following statistical tests are appropriate for analyzing the results of this study? Select all that apply.
- Paired t-testAnswer
- Wilcoxon signed-rank testAnswer
- CIndependent samples t-test
- DChi-square test
- EKruskal-Wallis test
Answer
The correct statistical tests for this scenario are the paired t-test (for normally distributed data) and the Wilcoxon signed-rank test (for non-normally distributed data).
The study evaluates a continuous variable (serum potassium) collected from a single group of patients at two distinct time points (baseline and four weeks later). Because the measurements are taken from the exact same individuals, the data is paired (dependent). The appropriate test for analyzing the difference in means for paired continuous data is the paired t-test, provided the differences are normally distributed. If the distribution is skewed or non-normal, the correct non-parametric alternative is the Wilcoxon signed-rank test.
Step-by-Step Solution
Key Concept
Selecting statistical tests for paired continuous variables