Name a nonparametric alternative to the paired samples t-test.

• A. Wilcoxon signed-rank test.
• B. Mann-Whitney U test.
• C. Independent samples t-test.
• D. ANOVA.

Answer: (A)

When you have paired or matched data and you can’t assume that the differences are normally distributed, a nonparametric alternative to the paired t-test is used. The Wilcoxon signed-rank test fits here because it assesses whether there is a systematic difference between the two measurements without relying on normality. It works by taking the differences between each pair, dropping any pairs with zero difference, ranking the absolute differences, and then using the signs of those differences to test whether the median difference is zero. Because it uses ranks instead of raw values, it doesn’t require normal distribution and is appropriate for ordinal data or skewed interval data. It’s more informative than a sign test since it incorporates the magnitude of differences, not just their direction. The other options are designed for independent samples or for parametric comparisons across groups, so they don’t serve as the direct nonparametric counterpart to a paired t-test.