Non-parametric Tests

Start with assumptions

Begin with the flashcards titled Non-parametric tests. They introduce assumptions, ask when to use a non-parametric test and contrast different kinds of data. The mind map on non-parametric significance tests offers another way into the topic. It covers situations such as samples that are not normally distributed and populations with unequal variances, as well as alternatives including data transformation.

Use the flashcards to check the terms, then return to the map to see how the ideas fit together.

Explore individual tests

The flashcards on non-parametric tests of differences discuss why Mann-Whitney U, Wilcoxon Signed Rank and Kruskal-Wallis are used, along with limitations. The Non Parametric Tests flashcards group tests by whether they are based on the binomial distribution or on ranks.

For a visual overview of a particular test, Kruskal Wallis ANOVA describes it as an alternative to ANOVA and shows the use of ranks. The Binomial Sign Test (S) flowchart presents criteria for use alongside a worked layout of participant data and differences.

Check your understanding

After studying a map or deck, use the quiz on non-parametric tests to practice questions about test comparisons. The quiz includes a question about what the Mann-Whitney test can substitute for. If a question is difficult, return to the relevant flashcards or map and study the terms before trying it again.

The quizzes on this page also include questions comparing non-parametric tests with their parametric counterparts when assumptions are met. They provide another way to review the distinctions introduced in the flashcards.

Place the topic in the unit

Non-parametric Tests is a section of Common Statistical Tests (t-tests, Chi-Square, Non-parametric) in Statistics and probability. The resources here give you a focused starting point within that unit, using several formats to explore test use and comparisons. Choose a map or flashcard deck to begin.