For a study comparing two independent samples with nonparametric data, which test is MOST appropriate?

• A. Mann-Whitney U Test
• B. Analysis of Variance
• C. Student's T-Test
• D. Fisher's Exact Test

Answer: (A)

When you’re comparing two independent groups and the data are nonparametric or ordinal, you need a test that doesn’t assume normal distribution. The Mann-Whitney U test fits this need by ranking all observations from both groups together and then comparing the sum of ranks between the groups. Because it uses ranks rather than actual values, it doesn’t rely on normality and is less sensitive to outliers. It’s effectively the nonparametric counterpart to the independent-samples t-test, so it’s ideal for detecting a difference in central tendency when you can’t assume normality.

If the p-value is small, you conclude there’s a difference in the distributions between the two groups, often interpreted as a difference in medians when the shapes of the distributions are similar. The other options rely on assumptions you don’t have here: Analysis of Variance and the Student’s t-test require normally distributed data (and t-tests assume equal variances for the two groups); they compare means and aren’t appropriate for nonparametric data. Fisher’s exact test is for categorical data in small samples, such as a 2x2 table, not for comparing ordinal or continuous outcomes between two groups.