UNIVERSITY OF NEGROS OCCIDENTAL - RECOLETOS

GRADUATE SCHOOL

First Semester 2022 - 2023

Chapter 5. Non-Parametric Test
According to LaMorte (2017), the nonparametric tests are sometimes called distribution-free tests because they
are based on fewer assumptions (e.g., they do not assume that the outcome is approximately normally distributed).
Parametric tests involve specific probability distributions (e.g., the normal distribution) and the tests involve estimation
of the key parameters of that distribution (e.g., the mean or difference in means) from the sample data.
In addition, the cost of fewer assumptions is that nonparametric tests are generally less powerful than their
parametric counterparts (i.e., when the alternative is true, they may be less likely to reject H 0). There are some
situations when it is clear that the outcome does not follow a normal distribution. These include situations:
● when the outcome is an ordinal variable or a rank,

● when there are definite outliers or

● when the outcome has clear limits of detection.

Nonparametric test and its Parametric counterpart

Wilcoxon Singed Rank

The Wilcoxon signed-rank test is the nonparametric test equivalent to the dependent t-test. As the Wilcoxon
signed-rank test does not assume normality in the data, it can be used when this assumption has been violated and the
use of the dependent t-test is inappropriate. It is used to compare two sets of scores that come from the same
participants (Leard Statistics)

Example:
The table below shows the hours of relief provided by two analgesic drugs in 12 patients suffering from arthritis.
Is there any evidence that one drug provides longer relief than the other? (use 0.01 level of significance)

Patients Drug 1 Drug 2

1 2.0 3.5

2 3.6 5.7

3 2.6 2.9

4 2.6 2.4

5 7.3 9.9

6 3.4 3.3

7 14.9 16.7
 8 6.6 6.0

9 2.3 3.8

10 2.0 4.0

11 6.8 9.1

12 8.5 20.9

Using 5 steps of hypothesis testing

1. State null and alternative hypothesis
Ho: There is no evidence that the one drug provides longer relief than the other.
H1: There is evidence that the one drug provides longer relief than the other.

2. Level of significance.
α = 0.01 level of significance.

3. Statistical Tool
Wilcoxon Singed Rank

4. Computation

Using JASP
1. Open the data, select T-test and click paired t-test.

2. Direct the variables Drug 1 and 2 to Variables Pairs.

 3. Click Wilcoxon signed-rank and Descriptives. Then go to Results.

Results

Paired Samples T-Test

Measure 1   Measure 2 Test Statistic df p
-
Drug 1 Drug 2 Student -2.168 11 0.053
Wilcoxon 7.000 0.013

Descriptives
  N Mean SD SE
Drug 1 12 5.217 3.839 1.108
Drug 2 12 7.350 5.912 1.707

5. Making decisions and conclusions

Based from the results, there was no enough evidence that the one drug provides longer relief than the other

[W = 7.00, p = 0.013] at 0.01 level of significance. This implies that the length of relief given by Drug 1 dose not differ

from Drug 2.

Activity

A firm wants to study the effect of music on the productivity of employees. One department of a certain factory
is selected at random to receive piped-in music for 30 days. There are 10 employees in the department. The following
table shows the average daily output for 30 days before the introduction of music and the average daily output for the
30 days during which music is piped into the department. Is there significant difference in the productivity of the
employees?

Before music During music

90 99
 80 85

92 98

85 83

81 88

85 99

72 80

85 91

70 80

88 94

Mann-Whitney U-test
The Mann-Whitney U test is used to compare differences between two independent groups when the
dependent variable is either ordinal or continuous, but not normally distributed. For example, you could use the Mann-
Whitney U test to understand whether attitudes towards pay discrimination, where attitudes are measured on an
ordinal scale, differ based on gender (i.e., your dependent variable would be "attitudes towards pay discrimination" and
your independent variable would be "gender", which has two groups: "male" and "female") (Leard Statistics).
Alternately, you could use the Mann-Whitney U test to understand whether salaries, measured on a continuous
scale, differed based on educational level (i.e., your dependent variable would be "salary" and your independent
variable would be "educational level", which has two groups: "high school" and "university"). The Mann-Whitney U test
is often considered the nonparametric alternative to the independent t-test although this is not always the case.

Example: A researcher gives a random sample of 15 college men and an independent random sample of 20 college
women a test to measure their knowledge of ecological issues. The table of scores shows, if there is significant
difference in the level of the participants knowledge on ecological issues when grouped according to sex? (use 0.05 level
of significance)

Sex Scores

Male 18.50

Male 14.00

Male 20.00

Male 19.00
 Male 19.50

Male 17.00

Male 16.00

Male 12.50

Male 12.00

Male 10.00

Male 12.40

Male 15.20

Male 12.50

Male 19.25

Male 11.00

Female 25.00

Female 23.00

Female 16.20

Female 19.75

Female 20.00

Female 19.10

Female 18.75

Female 21.10

Female 17.50

Female 17.75

Female 15.00

Female 21.00

Female 18.50

Female 17.25

Female 16.30

Female 18.00

Female 18.25

Female 24.00

Female 18.30

Female 19.20

Using 5 steps of hypothesis testing

1. State null and alternative hypothesis

Ho: There is no significant difference in the level of the participants knowledge on ecological issues when
grouped according to sex.
H1: There is significant difference in the level of the participants knowledge on ecological issues when
 grouped according to sex.

2. Level of significance
α = 0.05

3. Statistical Tool
Mann-Whetney U – test

4. Computation
Using JASP

1. Open the data, select T-test and click Independent sample t-test

2. Direct Sex to Grouping Variable and Score to Dependent Variables. Then Click Mann-Whitney and
Descriptives.

3. Go to Results

Results

Independent Samples T-Test

Test Statistic df p
Score Student 3.875 33 < .001
  Mann-Whitney 234.000 0.005
 Independent Samples T-Test
Test Statistic df p

Group Descriptives
  Group N Mean SD SE
Score Female 20 19.198 2.581 0.577
  Male 15 15.257 3.442 0.889

5. Making decisions and conclusion

Based from the result, there was significant difference in the level of the participants knowledge on ecological

issues when grouped according to sex [U = 234.00, p = 0.005] at 0.05 level of significance. This implies that women

college students perform better on ecological issues than men college students.

Activity
A firm wishes to compare two methods of communicating information about a new product. Two groups of
subjects are chosen to take part in the experiment. Subjects in the first group learn about the new product by Method A.
Subjects in the second group learn about it by Method B. At the end of the experiment each subject is given a test to
measure knowledge of the new product. The results are shown in the following table. Is there significant difference in
the knowledge of the new product when the subjects are grouped according to methods? (α = 0.05)

Methods Knowledge

Method A 50

Method A 59

Method A 60

Method A 71

Method A 80

Method A 81

Method A 80

Method A 78

Method A 72

Method A 77

Method A 73

Method B 53

Method B 54

Method B 58

Method B 78

Method B 65

Method B 69
 Method B 61

Method B 60

Method B 72

Method B 60

Method B 59

Kruskal-Wallis H-test
The Kruskal‐Wallis (Kruskal & Wallis, 1952) is a nonparametric statistical test that assesses the differences
among three or more independently sampled groups on a single, non‐normally distributed continuous variable. Non‐
normally distributed data (e.g., ordinal or rank data) are suitable for the Kruskal‐Wallis test.
In contrast, the one‐way analysis of variance (ANOVA), which is a parametric test, may be used for a normally
distributed continuous variable. The Kruskal‐Wallis test is an extension of the two‐group Mann‐Whitney U (Wilcoxon
rank) test. Thus, the Kruskal‐Wallis is a more generalized form of the Mann‐Whitney U test and is the nonparametric
version of the one‐way ANOVA.

Example:
The Department of Agriculture which to know whether they can conclude that three types of fertilizer have
different effects on the mean yield in bushels per acre of a certain grain. Each of the three types of fertilizer is applied to
four one-acre plots of ground. These plots are as alike with respect to relevant variables as possible. The plots are all
treated alike during the growing season. The table shows the yields of the 30 plots. Is there significant difference in the
mean yield of plots when grouped according to types of fertilizers? (α = 0.01)

Fertilizer Yields

Fertilizer A 45

Fertilizer A 40

Fertilizer A 41

Fertilizer A 46

Fertilizer A 45

Fertilizer A 50

Fertilizer A 47

Fertilizer A 41

Fertilizer A 43

Fertilizer A 44

Fertilizer B 42

Fertilizer B 44

Fertilizer B 43

Fertilizer B 47

Fertilizer B 45

Fertilizer B 47

Fertilizer B 48

Fertilizer B 50

Fertilizer B 51

Fertilizer B 49
 Fertilizer C 53

Fertilizer C 56

Fertilizer C 54

Fertilizer C 55

Fertilizer C 53

Fertilizer C 54

Fertilizer C 56

Fertilizer C 55

Fertilizer C 52

Fertilizer C 53

Using 5 Steps of hypothesis testing

1. State null and alternative hypothesis

Ho: There is no significant difference in the mean yield of plots when grouped according to types of
Fertilizers.
H1: There is significant difference in the mean yield of plots when grouped according to types of
Fertilizers.

2. Level of significance
α = 0.01

3. Statistical Tool
Kruskal – Wallis H-test

4. Computation
Using JASP

1. Open data, select ANOVA and click ANOVA

2. Direct Fertilizer to Fixed Factors and Yeilds to Dependent Variable, and click Discriptives
 3. Select Post Hoc Test, direct Fertilizer to the other box and click Dunn (Method)

4. Select Non-Parametric and direct Kruskal – Wallis to the other box

5. Results

Results

Kruskal-Wallis Test

Kruskal-Wallis Test
Factor Statistic df p
Fertilizer 20.635 2 < .001

Dunn's Post Hoc Comparisons - Fertilizer

Comparison z Wi Wj p p bonf p holm
A-B -1.095 8.350 12.650 0.137 0.411 0.137
 Dunn's Post Hoc Comparisons - Fertilizer
Comparison z Wi Wj p p bonf p holm
A-C -4.365 8.350 25.500 < .001 < .001 < .001
B-C -3.271 12.650 25.500 < .001 0.002 0.001

Descriptives - Yields
Fertilizer Mean SD N
A 44.200 3.084 10
B 46.600 3.026 10
C 54.100 1.370 10

5. Making decision and conclusion

Based from the results, there was significant difference in the mean yield of plots when grouped according to

types of Fertilizers [H(2) = 20.635, p < 0.001] at 0.05 level of significance. Utilizing Dunn Method, Fertilizer C creates

difference with Fertilizer B and A with z – value difference of 3.271 and 4.365 respectively. This implies that Fertilizer C

yields more product of grain than the other Fertilizers.

Activity
Incoming trainees in a large company are interviewed by 3 different interviewers with regard to their
management potential. Each interviewer gives the trainee a score from 0 to 100. The Personnel Manager wishes to
know if the 3 interviewers give consistent results. Seven (7) trainees are graded by the 3 interviewers with the following
results:
Interviewer Scores

Interview 1 71

Interview 1 84

Interview 1 60

Interview 1 78

Interview 1 65

Interview 1 53

Interview 1 72

Interview 2 94

Interview 2 68
 Interview 2 72

Interview 2 66

Interview 2 85

Interview 2 47

Interview 2 75

Interview 3 70

Interview 3 71

Interview 3 54

Interview 3 57

Interview 3 78

Interview 3 45

Interview 3 53

Is there significant differece on the interviers scores? (α = 0.01)

References:

Kruskal & Wallis, (1952), Kruskal-Wallis H-test,

https://onlinelibrary.wiley.com/doi/abs/10.1002/9780470479216.corpsy0491, Retrieved: May 9, 2020, Saturday
LaMorte, W (2017), When to Use a Nonparametric Test,
http://sphweb.bumc.bu.edu/otlt/MPH-Modules/BS/BS704_Nonparametric/BS704_Nonparametric2.html, Retrieved:
May 9, 2020, Saturday

Leard Statistics, Mann-Whitney U-test, https://statistics.laerd.com/spss-tutorials/mann-whitney-u-test-using-spss-

statistics.php, Retrieved May 9, 2020, Saturday.

Leard Statistics, Wilconxon Singed Rank, https://statistics.laerd.com/spss-tutorials/wilcoxon-signed-rank-test-using-spss-

statistics.php, Retrieved May 9, 2020, Saturday.