WILCOXON SIGNED RANK TEST

Submitted to:
Mr. Chuda Dhakal
Department of Statistics
Associate Professor
Institute of Agriculture and Animal Science
Kirtipur, Kathmandu, Nepal

Submitted by:
R-2019-ANU-01M Bandhana Regmi
R-2019-ANU-02M Santosh Kandel
R-2019-ANU-03M Kamana Adhikari
R-2019-ANU- 04M Sujata Aryal
R-2019-ANU-05M Shiva Poudel
R-2019-ANU-06M Ambeeka Bhattarai
R-2019-ANU-07M Babi Basnet
R-2019-ANU-08M Kushal Acharya
Department of Animal Nutrition & Fodder Production
Institute of Agriculture and Animal Science
Post Graduate Program
Kirtipur, Kathmandu, Nepal

March, 2021

1
Table of Content

s
ABSTRACT....................................................................................................................................3
1.INTRODUCTION.....................................................................................................................4-5
1.1.The History of Wilcoxon test..............................................................................................4
1.2.The test assumptions……………………………………………………………………..5
2.OBJECTIVES...............................................................................................................................6
3.METHODOLOGIES....................................................................................................................6
4.LITERATURE REVIEW........................................................................................................6-13
4.1.Application...........................................................................................................................7-8
4.1.1.When the Wilcoxon test is applicable..............................................................................7
4.1.2. When the Wilcoxon test is not applicable.......................................................................8
4.2.The Wilcoxon test formula......................................................................................................9
4.3.Examples of Wilcoxon test.................................................................................................9-13
5.RESULT……………………………………………………………………………………….
14
6.LIMITATION…………………………………….…………………………………………. ..14
7.CONCLUSION……………………………………………………………………………….. 14
8.REFRENCES…………………………………………………………………………………. 15

2
ABSTRACT
The Wilcoxon signed-rank test refers to a non-parametric statistical theory that is very significant
in carrying out tests of two related models as well as repeated dimensions on individual samples
to establish whether there are variations in their populace mean ranks. The analysis plays an
alternative means of assessment to the paired Student’s test for corresponding pairs as well as the
t-test for independent samples in the event that the populace is not normally distributed.

3
1.INTRODUCTION
A Wilcoxon signed-rank test is a nonparametric test that can be used to determine whether two
dependent samples were selected from populations having the same distribution.The Wilcoxon
signed-rank test is 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 (i.e. it is a paired difference test).
The Wilcoxon signed-rank can be used in the place of student paired t-test particularly when the
sample means are small. The assumptions made while using the Wilcoxon signed-rank test
enables it to be more accurate than the dependent variables t-test particularly when the sample
means are small. However, when the paired sample means are large, the t-test for non-dependent
variable would be appropriate.

The Wilcoxon signed-rank test refers to a non-parametric statistical theory that is very significant
in carrying out tests of two related models as well as repeated dimensions on individual samples
to establish whether there are variations in their populace mean ranks (Gravetter & Wallnau,
2009). Further, the test is also useful in assessing the differences existing between the population
mean ranks of matched samples.

In carrying out the Wilcoxon test, the statistics from the corresponding population are paired off.
The test also applies random sampling of the independent pairs (Gravetter & Wallnau, 2009).

Moreover, an ordinal scale is vital in measuring the statistics following a normal distribution. In
essence, the hypothesis testing of non-parametric data is essential in assessing records that can be
placed in a given order but lack the statistical figures. In fact, the test is invaluable in analyzing
clientele fulfillment (Gravetter & Wallnau, 2009).

1.1.The history of the Wilcoxon test

The test is named for Frank Wilcoxon (1892–1965) who, in a single paper, proposed both it and
the rank-sum test for two independent samples (Wilcoxon, 1945). The test was popularized
by Sidney Siegel (1956) in his influential textbook on non-parametric statistics. [3] Siegel used the
symbol T for a value related to, but not the same as, W. In consequence, the test is sometimes
referred to as the Wilcoxon T test, and the test statistic is reported as a value of T.

4
1.2.The test assumptions

For the Wilcoxon test to be appropriate, several assumptions are applied (Kirk, 2006). The first
assumption is that the data must be matched and have to be drawn from the corresponding
populace. Secondly, independent and random pairing must be attained. The final assumption is
that the data has to be ordinal. However, the normal distribution of data is necessary but not a
must requirement (Kirk, 2006).

Assumptions:

1.Independence:
The Wilcoxon sign test assumes independence, meaning that the paired observations are
randomly and independently drawn.

2.Dependent samples:
the two samples need to be dependent observations of the cases. The Wilcoxon sign test assess
for differences between a before and after measurement, while accounting for individual
differences in the baseline.

3.Continuous dependent variable:

Although the Wilcoxon signed rank test ranks the differences according to their size and is
therefore a non-parametric test, it assumes that the measurements are continuous in theoretical
nature. To account for the fact that in most cases the dependent variable is binomially
distributed, a continuity correction is applied.

5
2.OBJECTIVES
• We use the Wilcoxon signed-rank test to determine if two dependent samples are selected
from populations having the same distribution.
• We use the Wilcoxon rank sum test to determine if two independent sample are selected
from populations having the same distribution.

3.METHODOLOGY

This term paper was mainly prepared by collecting informations and essential materials.
Materials were collected from different journals, leaflets, booklets published by different
agriculture related organizations. Further informations were collected from the related websites.
These information were thoroughly reviewed and were incorporated in the review section.

4.LITERATURE REVIEW

A Wilcoxon signed-rank test is a nonparametric test that can be used to determine whether two
dependent samples were selected from populations having the same distribution. The test is
named for Frank Wilcoxon (1892–1965) who, in a single paper, proposed both it and the rank
sum- test for two independent samples (Wilcoxon, 1945). The test was popularized by Sidney
Siegel (1956) in his influential textbook on non-parametric statistics. Siegel used the symbol T
for a value related to, but not the same as .In consequence, the test is sometimes referred to as
the Wilcoxon T test, and the test statistic is reported as a value of T.

4.1 Application of Wilcoxon test

We use wilcoxon signed –rank test in two conditions:

When the sample cash is small (n≤15)
When sample size is small, a critical value against which to compare T can be found by table, to
determine whether the null hypothesis should be rejected. The critical value is located by using n
and ɑ.

6
If the observed value of T is less than or equal to the critical value of T, the decision is to reject
the null hypothesis.
When the sample cash is large (n>15)
For large samples, the T statistic is approximately normally distributed ans z score can be used as
the test statistic.

4.1.1.When the Wilcoxon test is applicable

The application of the Wilcoxon test is motivated by the improbability relating to the
assumptions of normality in the t-test. For instance, Wilcoxon test is applicable in the situations
that involve matched pairs as well as repeated dimension of similar items. Further, the Wilcoxon
test is applied in the case where there are improper mean variations in the t-test (Gravetter &
Wallnau, 2009).

In addition, the test remains indispensable in the circumstances when the distributions do not
obey the normal essential suppositions. For example, in the normal parametric statistical
procedures, the assumption is that the data will follow the uniform distributions. However, in the
actual circumstances, the data do not follow the uniform distribution. The test is also used when
the variables are at least ordinal.

In essence, the test is applied when similar participants carry out both situations for the study. In
other words, the test is suitable for scrutiny of information that originates from recurring
dimensional designs involving two situations. Further, the test is used when the statistics do not
convene to the necessities for a parametric test (Gravetter & Wallnau, 2009).

In this regard, the test is suitable when the data are not distributed normally as well as when there
are variations in marking the values of both the variances. In addition, for the Wilcoxon test to be
carried out, the statistics must be measured on an ordinal scale.

The sense behind conducting the test involves ranking the available figures to produce two order
sums for individual situations. The methodical variation between the circumstances leads to
placing the conditions with higher orders in one situation and the ones with lower ranks on the
other situation (Jackson, 2009). Further, if the ranks are equivalent, then the orders are
distributed in a similar array. The test is also significant in carrying out a single sample.

7
For example, in conducting assessment on shopping centers, the use of Wilcoxon test is highly
appropriate. In other words, in the study of several shops, the researcher would suppose that the
respondents are probable to back up or object to questions put to them with an equal likelihood
of half. In addition, when there is availability of various unrelated samples that require
comparisons, the application of the Wilcoxon rank test is vital.

4.1.2.When not to use the Wilcoxon test

As indicated above the Wilcoxon test is used in the situations where the sample population is
small. In the situations where the sampled population sizes are large, the Wilcoxon text cannot
be applied (Cleves, 2008).

In other words, while studying two population samples that are not identical or corresponding
and the samples are large then Wilcoxon test is not applicable. Instead, unpaired t-test is used.
The reason why the unpaired t-test is not preferable is that type 1 error is likely to be yielded
particularly in the circumstances when the population sample size is small (Cleves, 2008).

In other words, the unpaired t-test is preferable in the situations when the sample population size
is large. However, when the distribution is skewed and the sample population is large, then
Wilcoxon can still be applied (Martinez, 2007). In essence, Wilcoxon test yield better results
when the paired population sample being studied is small. When paired sample sizes are large,
then the Wilcoxon cannot be applied and instead the single t-test is used to test the variables.

According to Jackson (2009), Wilcoxon test can be used in the place of the paired student t-test.
Wilcoxon test is appropriate in the evaluation of data that are derived from the repeated
measures. In addition, in the circumstances that the derived data does not meet the parametric
test requirements or when the data are not distributed normally, then the Wilcoxon test is
applied.

8
4.2.The Wilcoxon test formula

As indicated, the Wilcoxon test assumes a formula that test the median of sample sizes of the
paired numbers. The sample size should be small. The hypothesis is whether there are variations
in the medians of the sample pairs (Martinez, 2007). In other words, the unacceptable premise
tests whether there is no existing deviations in the paired sample middle measure while the
acceptable premise tests otherwise.

To derive the formula N is understood to be the size of the sample or the amount of paired
samples. Therefore, the total data sample equals to 2N. If i=1, ….., N and x1i and x2i is to
represent the measurements, then Wilcoxon test (W) ={∑[sgn(x 2i-x1i)*Ri]}. In the
formula, sgn represent the sign function while Ri represent the ranks (Martinez, 2007). As
indicated, the hypothesis to be tested using the formula is as follows

 H0: there are no median differences between the pairs

 H1: median differences exist

4.3 Examples of Wilcoxon test

 How to Conduct the Wilcoxon Sign Test

9
The Wilcoxon signed rank test is the non-parametric of the dependent samples t-test.  Because
the dependent samples t-test analyzes if the average difference of two repeated measures is zero,
it requires metric (interval or ratio) and normally distributed data; the Wilcoxon sign test uses
ranked or ordinal data; thus, it is a common alternative to the dependent samples t-test when its
assumptions are not met.

The Wilcoxon signed rank test relies on the W-statistic.  For large samples with n>10 paired
observations the W-statistic approximates a normal distribution.  The W statistic is a non-
parametric test, thus it does not need multivariate normality in the data.

A research team wants to test whether a new teaching method increases the literacy of children. 
Therefore, the researchers take measure the literacy of 20 children before and after the teaching
method has been applied.  The literacy is measured on a scale from 0 to 10, with 10 indicating
high literacy.  The initial baseline shows an average literacy score of 5.9 and after the method
has been used the average increases to 7.6.

A dependent samples t-test cannot be used, as the distribution does not approximate a normal
distribution.  Also both measurements are not independent from each other; therefore, the Mann-
Whitney U-test cannot be used.

The first step of the Wilcoxon sign test is to calculate the differences of the repeated
measurements and to calculate the absolute differences.

10
The next step of the Wilcoxon sign test is to sign each rank.  If the original difference < 0 then
the rank is multiplied by -1; if the difference is positive the rank stays positive.

11
For the Wilcoxon signed rank test we can ignore cases where the difference is zero.  For all other
cases we assign their relative rank. In case of tied ranks the average rank is calculated.  That is if
rank 10 and 11 have the same observed differences both are assigned rank 10.5.
The next step of the Wilcoxon sign test is to sign each rank.  If the original difference < 0 then
the rank is multiplied by -1; if the difference is positive the rank stays positive.

12
The next step is to calculate the W+ and W-.
W+ = 4.5 + 4.5 = 9
W- = 1+ 2 + 4.5 + 4.5 + 7.5 + 7.5 + 9 + 10.5 + 10.5 + 12.5 + 12.5 + 14 + 17+ 17 + 17 + 17 + 17
= 181

We can double check this knowing 

Therefore,   and 
For large samples with n>10 the W-statistics approximates normal distribution,

with 
Further, we must make a compensation for ties, 4, 2, 2, 2, and 5, so we must reduce the variance

by 

13
14
5.RESULT
The shortcut to the hypothesis testing of the Wilcoxon signed rank-test is knowing the critical z-
value for a 95% confidence interval (or a 5% level of significance) which is z = 1.96 for a two-
tailed test and directionality.  Whenever a test is based the normal distribution the sample z value
needs to be 1.96 or higher to reject the null hypothesis.

6.LIMITATION
As demonstrated in the example, when the difference between the groups is zero, the
observations are discarded. This is of particular concern if the samples are taken from a discrete
distribution. In these scenarios the modification to the Wilcoxon test by Pratt 1959, provides an
alternative which incorporates the zero differences. This modification is more robust for data on
an ordinal scale.

7.CONCLUSION
The Wilcoxon signed-rank can be used in the place of student paired t-test particularly when the
sample means are small. The assumptions made while using the Wilcoxon signed-rank test
enables it to be more accurate than the dependent variables t-test particularly when the sample
means are small. However, when the paired sample means are large, the t-test for non-dependent
variable would be appropriate.

15
8.REFERENCES

1.  "Paired t–test - Handbook of Biological Statistics". www.biostathandbook.com.

Retrieved 2019-11-18.

2. Wilcoxon, Frank (Dec 1945). "Individual comparisons by ranking

methods" (PDF). Biometrics Bulletin. 1 (6): 80–
83. doi:10.2307/3001968. hdl:10338.dmlcz/135688. JSTOR 3001968.

3.Siegel, Sidney (1956). Non-parametric statistics for the behavioral sciences. New York:
McGraw-Hill. pp. 75–83. ISBN 9780070573482.

4.  Lowry,Richard. "Concepts & Applications of Inferential Statistics". Retrieved 5

November 2018.

5.  Lowry, Richard. "Concepts & Applications of Inferential Statistics". Retrieved 2020-12-17

6.Cleves, M. A. (2008). An introduction to survival analysis using stata. New York, NY: Stata
Press.

7.Gravetter, F. J. & Wallnau, L. B. (2009). Statistics for the behavioral sciences. Belmont, CA:
Cengage Learning.

8.Jackson, S. L. (2009). Statistics plain and simple. Belmont, CA: Cengage Learning.

9.Kirk, R. E. (2006). Statistics: An introduction. Belmont, CA: Cengage Learning.

10.Martinez, R. (2007). Diagnostics for choosing between Log-rank and Wilcoxon tests.

Michigan, MI: ProQuest.

16
17