Non-parametric tests are appropriate when data are which of the following?

Non-parametric tests are appropriate when data are which of the following?

• A. Data are normally distributed at interval level
• B. Data violate normality or are ordinal
• C. The sample size is large
• D. Variances are equal

Answer: (B)

Non-parametric tests are used when the assumptions underlying parametric tests aren’t met, most notably when the data are not normally distributed or when the data are ordinal rather than interval/ratio. Parametric tests rely on normality and on data measured on at least an interval scale with consistent variances; when these conditions fail, the results of parametric tests can be unreliable. Non-parametric methods don’t depend on the normal distribution and can handle ordinal data, making them appropriate in these situations. So, data that violate normality or are ordinal is the scenario for which non-parametric tests are best suited. If data are normally distributed on an interval scale, parametric tests are typically more powerful; large samples or equal variances don’t by themselves require non-parametric methods.

Non-parametric tests are used when the assumptions underlying parametric tests aren’t met, most notably when the data are not normally distributed or when the data are ordinal rather than interval/ratio. Parametric tests rely on normality and on data measured on at least an interval scale with consistent variances; when these conditions fail, the results of parametric tests can be unreliable. Non-parametric methods don’t depend on the normal distribution and can handle ordinal data, making them appropriate in these situations. So, data that violate normality or are ordinal is the scenario for which non-parametric tests are best suited. If data are normally distributed on an interval scale, parametric tests are typically more powerful; large samples or equal variances don’t by themselves require non-parametric methods.