What is the difference between the Wilcoxon signed-rank test and the Wilcoxon rank-sum test?

The signed-rank test is for paired or matched data (e.g., pre-test/post-test on the same subjects). The rank-sum test is for two independent samples (and is mathematically equivalent to the Mann–Whitney U test).

When should I use a Wilcoxon test instead of a t-test?

Use a Wilcoxon test when your data violates the normality assumption of the t-test, when you have ordinal data (ranks), or when your sample size is too small to reliably assess normality.

Is the Wilcoxon test a test of medians?

It is often described as a test of medians, but strictly, it tests whether the distribution of differences is symmetric around zero (for signed-rank) or whether one distribution is stochastically greater than the other (for rank-sum). It is a test of population mean ranks.

Do I need to report a test statistic for the Wilcoxon test?

Yes, standard reporting includes the test statistic (W for signed-rank, U or W for rank-sum), the sample sizes, and the p-value. Reporting effect size (e.g., r) is also recommended.

wilcoxon test

Low (specialist term in statistics)

UK/ˈwɪlkɒksən tɛst/US/ˈwɪlkɑːksən tɛst/

Formal, Academic, Technical (Statistics/Research)

My Flashcards

Definition

Meaning

A non-parametric statistical hypothesis test used to compare two related samples, matched samples, or repeated measurements on a single sample to assess whether their population mean ranks differ.

Specifically refers to two distinct tests developed by Frank Wilcoxon: the Wilcoxon signed-rank test (for paired data) and the Wilcoxon rank-sum test (for independent samples, equivalent to the Mann–Whitney U test). It’s used when the assumptions of parametric tests (like the t-test) are not met, typically due to non-normal distribution or ordinal data.

Linguistics

Semantic Notes

The name is capitalized as it derives from a proper name (Frank Wilcoxon). It is almost exclusively used in scientific and statistical contexts. Users must often specify which of the two tests (signed-rank or rank-sum) is being referred to.

Dialectal Variation

British vs American Usage

Differences

No significant difference in meaning or usage. Spelling remains consistent.

Connotations

Purely technical; carries no regional connotative differences.

Frequency

Equally low-frequency in both academic communities.

Vocabulary

Collocations

strong

perform a Wilcoxon testthe Wilcoxon signed-rank testthe Wilcoxon rank-sum testa significant Wilcoxon test result

medium

apply the Wilcoxon testresults from the Wilcoxon testWilcoxon test statisticpaired Wilcoxon test

weak

Wilcoxon test analysisusing a Wilcoxon testWilcoxon test outputWilcoxon test assumptions

Grammar

Valency Patterns

to test [something] using a Wilcoxon testto analyse [data] with a Wilcoxon signed-rank testto compare [two groups] via a Wilcoxon rank-sum test

Vocabulary

Synonyms

Neutral

Mann–Whitney U test (for rank-sum)non-parametric paired test (for signed-rank)

Weak

distribution-free test

Vocabulary

Antonyms

parametric testt-testANOVA

Usage

Context Usage

Business

Rare. Might appear in data analytics or market research reports comparing non-normal customer satisfaction scores.

Academic

Primary context. Used in methodology sections of psychology, medicine, biology, and social science papers.

Everyday

Virtually never used.

Technical

Core usage in statistics, data science, and research methodology.

Examples

By Part of Speech

verb

British English

We will Wilcoxon-test the paired observations.
The data were Wilcoxon-tested for significance.

American English

We need to Wilcoxon test these matched samples.
They Wilcoxon-tested the results.

adjective

British English

The Wilcoxon test results were non-significant.
A Wilcoxon test approach was deemed appropriate.

American English

The Wilcoxon test output is in Table 2.
We used a Wilcoxon test procedure.

Examples

By CEFR Level

The scientist used a test called the Wilcoxon test.
This test is useful when data is not normally distributed.

As the pre-test and post-test scores were not normal, we analysed them using the Wilcoxon signed-rank test.
The Wilcoxon rank-sum test showed a significant difference between the control and experimental groups.

Given the ordinal nature of the Likert-scale data and its violation of parametric assumptions, a Wilcoxon signed-rank test was employed to compare the median responses before and after the intervention.

Learning

Memory Aids

Mnemonic

Think: 'WILL COME and eXamine' ranks when data is NOT normal. 'Wil' (will) 'cox' (coxswain/ranks) 'on' test.

Conceptual Metaphor

A RANKING JUDGE (it judges based on the order/rank of data points, not their precise values).

Watch out

Common Pitfalls

Translation Traps (for Russian speakers)

Avoid translating as 'вилкоксон тест' in informal contexts; it's a direct borrowing. Confusion may arise with the similar-sounding 'критерий Вилкоксона' which is correct.

Common Mistakes

Uncapitalized ('wilcoxon test')
Using it without specifying 'signed-rank' or 'rank-sum'
Applying it to independent samples when referring to the signed-rank test.
Mispronouncing /ˈwɪlkɒksən/ as /waɪlˈkɒksən/.

Practice

Quiz

Fill in the gap

When your data is ordinal or not normally distributed, you might use a instead of a paired t-test.

Multiple Choice

Which of the following scenarios is appropriate for the Wilcoxon signed-rank test?

FAQ

Frequently Asked Questions

: The signed-rank test is for paired or matched data (e.g., pre-test/post-test on the same subjects). The rank-sum test is for two independent samples (and is mathematically equivalent to the Mann–Whitney U test).
: Use a Wilcoxon test when your data violates the normality assumption of the t-test, when you have ordinal data (ranks), or when your sample size is too small to reliably assess normality.
: It is often described as a test of medians, but strictly, it tests whether the distribution of differences is symmetric around zero (for signed-rank) or whether one distribution is stochastically greater than the other (for rank-sum). It is a test of population mean ranks.
: Yes, standard reporting includes the test statistic (W for signed-rank, U or W for rank-sum), the sample sizes, and the p-value. Reporting effect size (e.g., r) is also recommended.