Statistics
Textbooks
Boundless Statistics
Other Hypothesis Tests
Tests for Ranked Data
Statistics Textbooks Boundless Statistics Other Hypothesis Tests Tests for Ranked Data
Statistics Textbooks Boundless Statistics Other Hypothesis Tests
Statistics Textbooks Boundless Statistics
Statistics Textbooks
Statistics
Concept Version 7
Created by Boundless

Wilcoxon t-Test

The Wilcoxon ttt-test assesses whether population mean ranks differ for two related samples, matched samples, or repeated measurements on a single sample.

Learning Objective

  • Break down the procedure for the Wilcoxon signed-rank t-test.


Key Points

    • The Wilcoxon ttt-test can be used as an alternative to the paired Student's ttt-test, ttt-test for matched pairs, or the ttt-test for dependent samples when the population cannot be assumed to be normally distributed.
    • The test is named for Frank Wilcoxon who (in a single paper) proposed both the rank ttt-test and the rank-sum test for two independent samples.
    • The test assumes that data are paired and come from the same population, each pair is chosen randomly and independent and the data are measured at least on an ordinal scale, but need not be normal.

Terms

  • Wilcoxon t-test

    A non-parametric statistical hypothesis test used when comparing two related samples, matched samples, or repeated measurements on a single sample to assess whether their population mean ranks differ (i.e., it is a paired-difference test).

  • tie

    One or more equal values or sets of equal values in the data set.


Full Text

The Wilcoxon signed-rank t-test is a non-parametric statistical hypothesis test used when comparing two related samples, matched samples, or repeated measurements on a single sample to assess whether their population mean ranks differ (i.e., it is a paired difference test). It can be used as an alternative to the paired Student's ttt-test, ttt-test for matched pairs, or the ttt-test for dependent samples when the population cannot be assumed to be normally distributed.

The test is named for Frank Wilcoxon who (in a single paper) proposed both the rank ttt-test and the rank-sum test for two independent samples. The test was popularized by Siegel in his influential text book on non-parametric statistics. Siegel used the symbol TTT for the value defined below as WWW. In consequence, the test is sometimes referred to as the Wilcoxon TTT-test, and the test statistic is reported as a value of TTT. Other names may include the "ttt-test for matched pairs" or the "ttt-test for dependent samples."

Assumptions

  1. Data are paired and come from the same population.
  2. Each pair is chosen randomly and independent.
  3. The data are measured at least on an ordinal scale, but need not be normal.

Test Procedure

Let NNN be the sample size, the number of pairs. Thus, there are a total of 2N2N2N data points. For i=1,⋯,Ni=1,\cdots,Ni=1,⋯,N, let x1,ix_{1,i}x​1,i​​ and x2,ix_{2,i}x​2,i​​ denote the measurements.

H0H_0H​0​​: The median difference between the pairs is zero.

H1H_1H​1​​: The median difference is not zero.

1. For i=1,⋯,Ni=1,\cdots,Ni=1,⋯,N, calculate ∣x2,i−x1,i∣\left| { x }_{ 2,i }-{ x }_{ 1,i } \right|∣x​2,i​​−x​1,i​​∣ and sgn(x2,i−x1,i)\text{sgn}\left( { x }_{ 2,i }-{ x }_{ 1,i } \right)sgn(x​2,i​​−x​1,i​​), where sgn\text{sgn}sgn is the sign function.

2. Exclude pairs with ∣x2,i−x1,i∣=0\left|{ x }_{ 2,i }-{ x }_{ 1,i } \right|=0∣x​2,i​​−x​1,i​​∣=0. Let NrN_rN​r​​ be the reduced sample size.

3. Order the remaining pairs from smallest absolute difference to largest absolute difference, ∣x2,i−x1,i∣\left| { x }_{ 2,i }-{ x }_{ 1,i } \right|∣x​2,i​​−x​1,i​​∣.

4. Rank the pairs, starting with the smallest as 1. Ties receive a rank equal to the average of the ranks they span. Let RiR_iR​i​​ denote the rank.

5. Calculate the test statistic WWW, the absolute value of the sum of the signed ranks:

W=∣∑(sgn(x2,i−x1,i)⋅Ri)∣W= \left| \sum \left(\text{sgn}(x_{2,i}-x_{1,i}) \cdot R_i \right) \right|W=∣∑(sgn(x​2,i​​−x​1,i​​)⋅R​i​​)∣

6. As NrN_rN​r​​ increases, the sampling distribution of WWW converges to a normal distribution. Thus, for Nr≥10N_r \geq 10N​r​​≥10, a zzz-score can be calculated as follows: 

z=W−0.5σWz=\dfrac{W-0.5}{\sigma_W}z=​σ​W​​​​W−0.5​​

where

σW=Nr(Nr+1)(2Nr+1)6\displaystyle{\sigma_W = \sqrt{\frac{N_r(N_r+1)(2N_r+1)}{6}}}σ​W​​=√​​6​​N​r​​(N​r​​+1)(2N​r​​+1)​​​​​

If z>zcriticalz > z_{\text{critical}}z>z​critical​​ then reject H0H_0H​0​​.

For Nr<10N_r < 10N​r​​<10, WWW is compared to a critical value from a reference table. If W≥Wcritical,NrW\ge { W }_{ \text{critical,}{ N }_{ r } }W≥W​critical,N​r​​​​ then reject H0H_0H​0​​.

Alternatively, a ppp-value can be calculated from enumeration of all possible combinations of WWW given NrN_rN​r​​.

[ edit ]
Edit this content
Prev Concept
Mann-Whitney U-Test
Kruskal-Wallis H-Test
Next Concept
Subjects
  • Accounting
  • Algebra
  • Art History
  • Biology
  • Business
  • Calculus
  • Chemistry
  • Communications
  • Economics
  • Finance
  • Management
  • Marketing
  • Microbiology
  • Physics
  • Physiology
  • Political Science
  • Psychology
  • Sociology
  • Statistics
  • U.S. History
  • World History
  • Writing

Except where noted, content and user contributions on this site are licensed under CC BY-SA 4.0 with attribution required.