What assumption underlies non-parametric tests?

What assumption underlies non-parametric tests?

• A. Not a normal distribution, distribution is skewed
• B. Normal distribution
• C. Homoscedasticity
• D. Equal sample sizes

Answer: (A)

Non-parametric tests are used when we cannot assume a normal distribution. They’re distribution-free and often rely on ranks rather than raw scores, which makes them well-suited for skewed data or when measurements are ordinal. The key requirement is that the observations be independent and that the data are at least ordinal; normality, a hallmark of many parametric tests, is not assumed. Variances or sample sizes aren’t the defining constraints in these tests, which is why the option pointing to non-normal (skewed) distributions best captures what underlies non-parametric methods.

Non-parametric tests are used when we cannot assume a normal distribution. They’re distribution-free and often rely on ranks rather than raw scores, which makes them well-suited for skewed data or when measurements are ordinal. The key requirement is that the observations be independent and that the data are at least ordinal; normality, a hallmark of many parametric tests, is not assumed. Variances or sample sizes aren’t the defining constraints in these tests, which is why the option pointing to non-normal (skewed) distributions best captures what underlies non-parametric methods.