Lesson 7 of 7 · All lessons in this module
- The six steps and stating the hypotheses
- Type I and Type II errors, significance and powerVideo lesson
- Critical values, p-values and the decision ruleVideo lesson
- Testing a single mean
- Two means: independent vs paired samplesVideo lesson
- Testing variances: chi-square and FVideo lesson
- Parametric vs nonparametric tests
Lesson 7 of 7 · 12 min
Parametric vs nonparametric tests
Parametric tests are about parameters and lean on distribution assumptions; when those assumptions fail, the data are ranks, or the question is not about a parameter, use a nonparametric test.
In short
- A parametric test concerns a population parameter (mean, variance) and relies on specific distributional assumptions, usually normality.
- A nonparametric test is not concerned with a parameter, or makes only minimal assumptions about the population.
- Four reasons to go nonparametric: assumptions not met, outliers, data given as ranks (ordinal scale), or a hypothesis that is not about a parameter.
- Alternatives: single mean → Wilcoxon signed-rank; paired differences → Wilcoxon signed-rank or sign test; two independent means → Mann–Whitney U (Wilcoxon rank sum).
- When the parametric assumptions hold, the parametric test is preferred because it usually has more power.
Unlock this lesson free for 7 days
Create a free account to get 7 days free: every lesson, video and flashcard, and the question bank. No card needed.