Non - Parametric Test

Nonparametric tests are methods of statistical analysis that do not require a

distribution to meet the required assumptions to be analyzed (especially if the data
is not normally distributed). Due to this reason, they are sometimes referred to as
distribution-free tests. Nonparametric tests serve as an alternative to parametric
tests such as T-test or ANOVA that can be employed only if the underlying data
satisfies certain criteria and assumptions.

Specific objectives:

At the end of the lesson, students are expected to:

1. examine the difference between an observed sample distribution and an
assumed distribution in a population;
2. assess whether two related or dependent observations for treatments are
different or not;
3. describe when a non-parametric test for two independent samples be used;
4. determine the degree or extent to which the unrelated groups differ on
some variable;
5. evaluate differences among multiple distributions when the response
variable is dichotomous or ranks;
6. analyze differences in three or more independent groups;
7. interpret the results of the analysis; and
8. determine whether there is relationship between variables that are of rank
order.

Chi- Square (x2) Goodness of Fit Test

Chi-Square goodness of fit test is a non-parametric test that is used to find

out how the observed value of a given phenomenon is significantly different from
the expected value. In Chi-Square goodness of fit test, the term goodness of fit is
used to compare the observed sample distribution with the expected probability

101
distribution. Chi-Square goodness of fit test determines how well theoretical
distribution (such as normal, binomial, or Poisson) fits the empirical distribution. In
Chi-Square goodness of fit test, sample data is divided into intervals. Then the
numbers of points that fall into the interval are compared, with the expected
numbers of points in each interval.

Procedure for Chi-Square Goodness of Fit Test:

1. Ho: Assumes that there is no significant difference between the observed
and the expected value
H1: Assumes that there is a significant difference between the observed
and the expected value.
2. Test-statistic: Use x2 one sample test.
3. Level of significance: Use 𝛼 level of significance.
4. Decision Criterion:
Reject Ho if 𝒙𝟐𝒄 ≥ 𝒙𝟐𝒕(𝒏−𝟏)𝒅𝒇

Reject Ho if 𝒙𝟐𝒄 ≤ 𝒙𝟐𝒕(𝒏−𝟏)𝒅𝒇

Reject Ho if 𝒙𝟐𝒄 ≥ 𝒙𝟐𝜶(𝒏−𝟏)𝒅𝒇

𝟐

5. Compute: Ei = n ×𝝁𝒊
where Ei = expected value per row
n = sample size
𝜇𝑖 = assumed distribution
(𝑶− 𝑬)𝟐
x2 =∑ [ 𝑬
]

Where: x2= Chi-Square goodness of fit test

O= observed value
E= expected value

Degree of freedom:
In Chi-Square goodness of fit test, the degree of freedom depends on the
distribution of the sample. The following table shows the distribution and an
associated degree of freedom:

102
Type of distribution No of constraints Degree of freedom
Binominal distribution 1 n-1
Poisson distribution 2 n-2
Normal distribution 3 n-3

6. State your decision based on the decision criterion and the t-computed.
7. State your conclusion.

Example:
Table 3.2.1.a shows the distribution of various ethnic groups in the
population of Sultan Kudarat province based on the Philippine Statistics Authority
(PSA) census. Five years later a random sample of 2,500 residents of the province
was taken, with the results given in the table 3.2.1.b (along with the probability
distribution from the census year). Test at 1% level of significance whether there is
sufficient evidence in the sample to conclude that the distribution of ethnic groups
in the province five years after the census had changed from that in the census
year.

Table 3.2.1.a: Ethnic Groups in the Census Year

Hiligaynon/ Manobo/Ata
Ethnicity Ilocano Cebuano Others Not Reported
Ilonggo Manobo
Proportion 0.743 0.216 0.012 0.012 0.008 0.009

Table 3.2.1.b: Sample Data Five Years After the Census Year

Ethnicity Assumed Distribution Observed Frequency

Hiligaynon/Ilonggo 0.743 1,732
Ilocano 0.216 538
Cebuano 0.012 32
Manobo/Ata Manobo 0.012 42
Others 0.008 133
Not Reported 0.009 23

Solution:
1. Ho: The distribution of ethnic group has not changed.
H1: The distribution of ethnic group has changed.
2. Test-statistic: Use x2 one sample test.

103
3. Level of significance: Use 𝛼 = 0.01 level of significance.
4. Decision Criterion: Reject Ho if 𝒙𝟐𝒄 ≥ 𝒙𝟐𝒕(𝟔−𝟏)𝒅𝒇 = 15.09 (refer Appendix
Table B).
5. Computation:
Compute the expected number of each row using the formula Ei = n ×𝝁𝒊
with n – 2500.
E1 = 2500 × 0.743 = 1857
E2 = 2500 × 0.216 = 540
E3 = 2500 × 0.012 = 30
E4 = 2500 × 0.012 = 30
E5 = 2500 × 0.008 = 20
E6 = 2500 × 0.009 = 22.5

Observed and Expected Frequencies Five Years After the Census Year

Assumed Observed Expected

Ethnicity
Distribution Frequency Frequency
Hiligaynon/Ilonggo 0.743 1,732 1857
Ilocano 0.216 538 540
Cebuano 0.012 32 30
Manobo/Ata Manobo 0.012 42 30
Others 0.008 133 20
Not Reported 0.009 23 22.5
Total 100% 2500 2500

(𝑶− 𝑬)𝟐
x2 =∑ [ 𝑬
]

(𝟏𝟕𝟑𝟐−𝟏𝟖𝟓𝟕.𝟓)𝟐 (𝟓𝟑𝟖−𝟓𝟒𝟎)𝟐 (𝟑𝟐−𝟑𝟎)𝟐 (𝟒𝟐−𝟑𝟎)𝟐 (𝟏𝟑𝟑−𝟐𝟎)𝟐 (𝟐𝟖−𝟐𝟐.𝟓)𝟐

x2 =∑ [ + + + + + ]
𝟏𝟖𝟓𝟕 𝟓𝟒𝟎 𝟑𝟎 𝟑𝟎 𝟐𝟎 𝟐𝟐.𝟓

= 651.881

6. Decision: Since 𝒙𝟐𝒄 = 651.881 ≥ 𝒙𝟐𝒕(𝟓)𝒅𝒇 = 15.09, there is sufficient evidence

to reject Ho at 1% level of confidence.
7. Conclusion: Therefore, the ethnic distribution in Sultan Kudarat province
has changed in the five years since the PSA census.

104
Sign Test Wilcoxon Matched -Paired Signed-Ranks Test

The Wilcoxon Sign Test is used to determine whether the mean ranks of
two dependent, or matched, samples are different from each other. This is the
non-parametric alternative to the dependent sample t-test. Technically, it tests if
the average difference of a continuous variable that is measured twice is zero.
The test statistic for the Wilcoxon Signed Rank Test is W, defined as the
smaller of 𝑊+ (sum of the positive ranks) and 𝑊− (sum of the negative ranks). If
the null hypothesis is true, we expect to see similar numbers of lower and higher
ranks that are both positive and negative (i.e., 𝑊+ and 𝑊− would be similar). If
the research hypothesis is true, we expect to see more higher and positive ranks.
𝑛(𝑛+1)
If the number of observations/pairs 2
is greater than 20, use the normal

approximation.
In computing for the test statistic, we use the formula:
𝒏(𝒏−𝟏)
𝑾−
𝟒
Z= 𝒏(𝒏+𝟏)(𝟐𝒏+𝟏)
√
𝟐𝟒

where: W = the smaller sum of ranks

n = sample
𝒕𝟐 −𝒕
If there is tied ranks, reduce n by for each t tied ranks.
𝟒𝟖

Example:
The researcher assesses the effectiveness of a new drug designed to
reduce repetitive behaviors in children affected with autism. A total of 8 children
with autism enroll in the study and the amount of time that each child is engaged
in repetitive behavior during three hours observation periods are measured both
before treatment and then again after taking the new medication for a period of 1
week. The data are shown below.

Child Before Treatment After 1 week of Treatment

1 85 75
2 70 50
3 40 50
4 65 40
5 80 20
6 75 65

105
 7 55 40
8 20 25

Solution:

1. Ho: The treatment is not significantly effective in reducing repetitive behavior

in children with autism.

H1: The treatment is significantly effective in reducing repetitive behavior in

children with autism.

2. Test statistic: Use Wilcoxon Signed Rank Test for dependent sample
3. Level of significance: Use 𝛼 = 0.05
4. Decision Criterion: Reject Ho if 𝑤𝑐 ≤ 𝑤𝑡(0.05) = 6.
5. Computation:
First, solve for the difference between before treatment (a) and after
treatment (b)
Using the formula: 𝐷𝑖 = 𝑎 - b
Difference
After 1 week of
Child Before Treatment (a) (Before and
Treatment (b)
After) 𝑫𝒊
1 85 75 10
2 70 50 20
3 40 50 -10
4 65 40 25
5 80 20 60
6 75 65 10
7 55 40 15
8 20 25 -5

The next step is to rank the difference scores. We first order the absolute
values of the difference scores and assign rank from 1 through n to the smallest
through largest absolute values of the difference scores, and assign the mean rank
when there are ties in the absolute values of the difference scores.
Observed Ordered absolute Values of
Ranks
Difference Difference
10 -5 1
20 10 3
-10 -10 3
25 10 3
60 15 5

106
 10 20 6
15 25 7
-5 60 8

The final step is to attach the signs ("+" or "-") of the observed differences
to each rank as shown below.
Observed Ordered absolute Signed Ranks
Ranks
Difference Values of Difference
10 -5 1 -1
20 10 3 3
-10 -10 3 -3
25 10 3 3
60 15 5 5
10 20 6 6
15 25 7 7
-5 60 8 8

𝑊+ = 3 +3 +5 + 6 + 7 + 8 = 32

𝑊− = 1 + 3 = 4

𝒏𝟏 (𝒏𝟏 +𝟏) 𝟖(𝟖+𝟏)

N= = = 36, that is 32 + 4.
𝟐 𝟐

In this example, 𝑊+ = 32 and 𝑊− = 4. Recall that the sum of the ranks (ignoring
𝒏𝟏 (𝒏𝟏 +𝟏)
the signs) will always equal . As a check on our assignment of ranks, we
𝟐
𝒏𝟏 (𝒏𝟏 +𝟏) 𝟖(𝟖+𝟏)
have 𝟐
= 𝟐
= 36 which is equal to 32+4. The test statistic is W = 4. (see

Appendix Table H).

6. Decision: Since 𝑊𝑐 = 0< 𝑤𝑡(0.05) = 6, reject Ho.

7. Conclusion: Therefore, the treatment is significantly effective in reducing
repetitive behavior in children with autism.

Mann – Whitney U Test

A popular nonparametric test to compare outcomes between two

independent groups is the Mann Whitney U test. The Mann Whitney U test,

107
sometimes called the Mann Whitney Wilcoxon Test or the Wilcoxon Rank Sum
Test, is used to test whether two samples are likely to derive from the same
population (i.e., that the two populations have the same shape). Some
investigators interpret this test as comparing the medians between the two
populations.

Methods:
Case1. When the sample sizes are equal, 𝑛1 = 𝑛2 .
i. Put all observations in single array tagging each observation to differentiate
the origin of each observation.
ii. Rank the observations in the combined array.
iii. Assign the average rank in case of ties.
iv. Sum the rank of the first sample 𝑀1 and the rank of the second sample 𝑀2
and compute M = min (𝑀1 , 𝑀2 ).
v. Compare M with tabular value (Appendix Table B).
vi. Decision Criterion: Reject 𝐻𝑜 if M≤ 𝑀𝑡 .

Case 2. When sample sizes are unequal 𝑛1 < 𝑛2 .

i. Do step 1 to 3 of case 1.
ii. Find the total of the ranks for sample that has the smaller size 𝑛1 (𝑀1 ).
iii. Compute for 𝑀2 = 𝑛1 (𝑛1 + 𝑛2 + 1) − 𝑀1 ).
iv. Determine M with the tabular value.
v. Decision criterion: Reject 𝐻𝑜 if M≤ 𝑀𝑡 .

Example:

Consider a Phase II clinical trial designed to investigate the effectiveness of

a new drug to reduce symptoms of asthma in children. A total of n=10 participants
are randomized to receive either the new drug or a placebo. Participants are asked
to record the number of episodes of shortness of breath over a 1-week period
following receipt of the assigned treatment. The data are shown below.

Placebo (P) 7 5 6 4 12
New Drug (ND) 3 6 4 2 1

108
 Is there a difference in the number of episodes of shortness of breath over
a 1-week period in participants receiving the new drug as compared to those
receiving the placebo? By inspection, it appears that participants receiving the
placebo have more episodes of shortness of breath, but is this statistically
significant?

Solution:

1. Ho: The new drug has no effect in reducing symptoms of asthma.

H1: The new drug is effective in reducing symptoms of asthma.
2. Test-statistic: Use the Mann-Whitney test
3. Level of significance: Use 𝛼 = 0.05 level of significance.
4. Decision Criterion: Reject Ho if 𝑀𝑐 ≤ 𝑀𝑡(0.05) = 17(refer Appendix Table B).
5. Computation:
𝒏𝟏=𝟓, 𝒏𝟐=𝟓
ND ND ND P ND P ND P P P
Array 1 2 3 4 4 5 6 6 7 12
Rank 1 2 3 4.5 4.5 6 7.5 7.5 9 10

𝑀1 = 1 + 2 + 3 + 4.5 + 7.5 = 18
𝑀2 = 4.5 + 6 + 7.5 + 9 + 10 = 37
M = min(𝑀1 , 𝑀2 )
= min(18, 37)
= 18
6. Decision: Since 𝑀𝑐 = 18 > 𝑀𝑡 (0.05) = 17, accept 𝐻𝑜 .
7. Conclusion: Therefore, we can conclude that the new drug is not effective
in reducing symptoms of asthma.

X2 Test for k Independent Samples

The Chi-Square test of independence is used to determine if there is a
significant relationship between two nominal (categorical) variables. The
frequency of each category for one nominal variable is compared across the

109
categories of the second nominal variable. The data can be displayed in a
contingency table where each row represents a category for one variable and each
column represents a category for the other variable.
To calculate the chi-squared statistic, take the difference between a pair of
observed (O) and expected values (E), square the difference, and divide
that squared difference by the expected value. Repeat this process for all cells in
your contingency table and sum those values.
𝒓 𝒌
𝟐
(𝑶𝒊𝒋 − 𝑬𝒊𝒋 )𝟐
𝑿 = ∑ ∑
𝑬𝒊𝒋
𝒊=𝟏 𝒋=𝟏

If the agreement between the observed and expected frequencies is close,

the differences (𝑂𝑖𝑗 − 𝐸𝑖𝑗 ) will be small and the chi-square will be small. However,
if the divergence is large, the value of chi-square as computed in the equation will
also be large. Roughly speaking, the larger the value of chi-square, the less likely
it is that the observed frequencies came from the population on which the null
hypothesis and the expected frequencies are based.

Procedure:
First, calculate the expected value of the two nominal variables. We can calculate
the expected value of the two nominal variables by using this formula:
∑𝒄𝒌=𝟏 𝑶𝒊𝒋 ∑𝒓𝒌=𝟏 𝑶𝒌𝒋
𝑬𝒊𝒋 =
𝑵
Where
𝑬𝒊𝒋 = expected value
∑𝒄𝒌=𝟏 𝑶𝒊𝒋 = Sum of the ith column
∑𝒓𝒌=𝟏 𝑶𝒌𝒋 = Sum of the kth row
N = total number

After calculating the expected value, we will apply the following formula to calculate
the value of the Chi-Square test of Independence:
(𝑶𝒊𝒋 − 𝑬𝒊𝒋 )𝟐
𝑿𝟐 = ∑𝒓𝒊=𝟏 ∑𝒌𝒋=𝟏
𝑬𝒊𝒋

110
 = Chi-Square test of Independence
𝑶𝒊𝒋 = Observed value of two nominal variables
𝑬𝒊𝒋 = Expected value of two nominal variables

Degree of freedom is calculated by using the following formula:

DF = (r-1)(c-1)
Where: DF = Degree of freedom
r = number of rows
c = number of columns

Example:

Suppose you want to assess whether there is a relationship between

exercise on campus and students' living arrangements. As part of the survey,
graduates were asked where they lived their senior year. The response options
were dormitory, on-campus apartment, off-campus apartment, and at home (i.e.,
commuted to and from the university). The data are shown below.

No Regular Sporadic Regular

Total
Exercise Exercise Exercise
Dormitory 32 30 28 90
On-Campus
74 64 42 180
Apartment
Off-Campus
110 25 15 150
Apartment
At Home 39 6 5 50

Total 255 125 90 470

Based on the data, is there a relationship between exercise and student's

living arrangement?

Solution:

111
 1. Ho: There is no significant relationship between exercise and student’s living
arrangement.
H1: There is a significant relationship between exercise and student’s living
arrangement.
2. Test-statistic: Use the Chi square Test
3. Level of significance: Use 𝛼 = 0.05 level of significance.
4. Decision Criterion: Reject Ho if 𝑋𝑐2 ≥ 𝑋(2𝛼=.05)(3−1)(4−1)𝑑𝑓 = 12.59 (refer
Appendix Table B).

5. Computation:
(𝑅𝑜𝑤 𝑇𝑜𝑡𝑎𝑙 × 𝐶𝑜𝑙𝑢𝑚𝑛 𝑇𝑜𝑡𝑎𝑙)
Expected Frequency =
𝑁
90 ×255
= = 48.8
470

No Regular Sporadic Regular

Total
Exercise Exercise Exercise
Dormitory 32 30 28 90
Expected Frequency
48.8 23.9 17.2
On-Campus Apartment 74 64 42 180
Expected Frequency
97.7 47.9 34.5
Off-Campus Apartment 110 25 15 150
Expected Frequency
81.4 39.9 28.7
At Home 39 6 5 50
Expected Frequency
27.1 13.3 9.6

Total 255 125 90 470

The test statistic is computed as follows:

112
 6. Decision: Since 𝑋𝑐2 = 𝟔𝟎. 𝟓 > 𝑋(2𝛼=.05)(3−1)(4−1)𝑑𝑓 = 12.59, reject Ho.

7. Conclusion: Therefore, it is evident that there is no significant relationship

between exercise and student’s living arrangement.

Kruskal – Wallis Test One-way ANOVA by Ranks

The Kruskal-Wallis H test (sometimes also called the "one-way ANOVA on

ranks") is a rank-based nonparametric test that can be used to determine if there
are statistically significant differences between two or more groups of an
independent variable on a continuous or ordinal dependent variable. It is
considered the nonparametric alternative to the one-way ANOVA, and an
extension of the Mann-Whitney U test to allow the comparison of more than two
independent groups.
The Kruskal – Wallis Test – statistics is defined as
𝒌
𝟏𝟐 𝑹𝟐𝒋
𝑲𝑾 = [∑ ] − 𝟑(𝒏 + 𝟏)
𝒏(𝒏 + 𝟏) 𝒏𝒋
𝒋=𝟏

The null hypothesis is rejected if 𝐾𝑊 exceeds the critical value of Chi-square

at 𝛼 level of significance with degree of freedom to k – 1.

Example:
An opinion poll was given to students. Their opinions were expressed by
using Likert scale with 5 choices. (1) strongly disagree, (2) disagree, (3) neutral,
(4) agree, (5) strongly agree. A researcher claims that the distributions for student’s
response to three of the questions in the poll: Q1, Q2, and Q3 are the same. Ten
students’ responses were randomly selected. Their responses to three questions
are listed below. Let 𝛼 = 0.10, using the Kruskal-Wallis Test, can you reject the
researcher’s claim?

113
 Student Q1 Q2 Q3
1 3 4 4
2 4 4 5
3 4 4 4
4 4 4 4
5 4 3 3
6 5 3 5
7 3 2 3
8 2 3 3
9 3 3 2
10 3 3 4

Solution:
1. Ho: The responses of students in the three questions are the same.
H1: At least two of the students’ responses to three questions are not the
same.
2. Test-statistic: Use the Chi-square test
3. Level of significance: Use 𝛼 = 0.10 level of significance.
4. Decision Criterion: Reject Ho if 𝑲𝑾≤ 𝑲𝑾(.𝟏𝟎)(𝟑−𝟏)𝒅𝒇 = 𝟒. 𝟔𝟏(refer Appendix
Table B).
5. Computation:
a. Rank the combine samples from lowest to the highest
b. For each individual sample, find the number of observations and the
sum of the ranks.

Student Q1 Rank Q2 Rank Q3 Rank

1 3 9.5 4 22.36 4 22.36
2 4 22.36 5 28.5 5 28.5
3 4 22.36 4 22.36 4 22.36
4 4 22.36 4 22.36 4 22.36
5 4 22.36 3 9.5 3 9.5
6 5 28.5 3 9.5 5 28.5
7 3 9.5 2 2 3 9.5
8 2 2 3 9.5 3 9.5
9 3 9.5 3 9.5 2 2
10 3 9.5 3 9.5 4 22.36
𝑹𝒋 157.94 145.08 176.94
̅̅̅̅̅
𝑹_𝒋 15.794 14.508 17.694
𝒏𝒋 10 10 10

𝟏𝟐 𝑹𝟐𝒋
𝑲𝑾 = [∑𝒌𝒋=𝟏 𝒏 ] − 𝟑(𝒏 + 𝟏)
𝒏(𝒏+𝟏) 𝒋

114
 𝟏𝟐 𝟏𝟓𝟕.𝟗𝟒𝟐 𝟏𝟒𝟓.𝟎𝟖𝟐 𝟏𝟕𝟔.𝟗𝟒𝟐
= 𝟑𝟎(𝟑𝟎+𝟏) [ + + ] − 𝟑(𝟑𝟎 + 𝟏)
𝟏𝟎 𝟏𝟎 𝟏𝟎

= 0.01290 (2494.504 + 2104.821 + 3130.776) – 93

= 0.01290 (7730.101) – 93
= 6.718

6. Decision: Since 𝑲𝑾 = 𝟔. 𝟕𝟏𝟖 ≥ 𝑲𝑾(.𝟏𝟎)(𝟑−𝟏)𝒅𝒇= 4.61, there is no sufficient

evidence to reject Ho.
7. Conclusion: Therefore, the researcher’s claim on the students’ responses in
the three questions is true.

Kendall Rank Correlation Coefficient: r

Correlation is a bivariate analysis that measures the strength of association

between two variables and the direction of the relationship. In terms of the strength
of relationship, the value of the correlation coefficient varies between +1 and -1. A
value of ± 1 indicates a perfect degree of association between the two
variables. As the correlation coefficient value goes towards 0, the relationship
between the two variables will be weaker. The direction of the relationship is
indicated by the sign of the coefficient; a + sign indicates a positive relationship
and a – sign indicates a negative relationship. Usually, in statistics, we measure
four types of correlations: Pearson correlation, Kendall rank correlation, Spearman
correlation, and the Point-Biserial correlation.
The Kendall rank correlation coefficient or Kendall's tau statistic is used to
estimate a rank-based measure of association. This test may be used if the data
do not necessarily come from a bivariate normal distribution. Tau is the Kendall
correlation coefficient.
Kendall rank correlation is a non-parametric test that measures the strength
of dependence between two variables. If we consider two samples, a and b, where
each sample size is n, we know that the total number of pairings with a b
𝒏(𝒏−𝟏)
is . The following formula is used to calculate the value of Kendall rank
𝟐

correlation:

115
 𝒏𝒄 − 𝒏𝒅 𝑪−𝑫
𝝉= 𝒏(𝒏−𝟏) or Kendall tau = 𝑪 + 𝑫
𝟐

𝒏𝒄 = number of concordant 𝐶 = Concordant

𝒏𝒅 = Number of discordant 𝐷 = Discordant

Kendall’s correlation coefficient uses pairs of observations and determines

the strength of association based on the pattern on concordance and discordance
between the pairs.
Procedure:
1) Make a table of rankings. The rankings for Column A should be in
ascending order (from least to greatest).
2) Count the number of concordant pairs, using the Column C. Concordant
pairs are how many larger ranks are below a certain rank. If the following
rank is smaller, do not count the concordant.
3) Count the number of discordant pairs and insert them into the Column D.
The number of discordant pairs is similar to Step 2, only you’re looking for
smaller ranks, not larger ones.
4) Sum the values in the two columns (Column C and Column D).
5) Insert the total in to the formula:
𝒏𝒄 − 𝒏𝒅
𝝉=
𝒏(𝒏 − 𝟏)
𝟐

Example:
The architectural drafting students exhibits 8 artworks during University Day.
The artwork ranked by two committee: the faculty and administrators. Determine
the degree of relationship between the faculty and administrator’s rank.
Faculty Rank Admin. Rank Concordant Discordant
1 2 6 1
2 1 6 0
3 5 3 2
4 4 3 1
5 3 3 0
6 7 1 1
7 6 1 0
8 8
TOTAL 23 5

116
 𝑪−𝑫
Kendall tau = 𝑪 + 𝑫
𝟐𝟑−𝟓 𝟏𝟖
= 𝟐𝟑+𝟓 = 𝟐𝟖 = 0.6429

The resulting coefficient shows that there is a moderate degree of

association between the faculty and administrators’’ ranking given to artworks
exhibit.

117
Summary:

✓ Chi-Square goodness of fit test is a non-parametric test that is used to find out
how the observed value of a given phenomenon is significantly different from
the expected value. In Chi-Square goodness of fit test, the term goodness of
fit is used to compare the observed sample distribution with the expected
probability distribution.
✓ The Wilcoxon Sign Test is used to determine whether the mean ranks of two
dependent, or matched, samples are different from each other. This is the non-
parametric alternative to the dependent sample t-test.
✓ The Mann Whitney U test, sometimes called the Mann Whitney Wilcoxon Test
or the Wilcoxon Rank Sum Test, is used to test whether two samples are likely
to derive from the same population (i.e., that the two populations have the same
shape). Some investigators interpret this test as comparing the medians
between the two populations.
✓ The Chi-Square test of independence is used to determine if there is a
significant relationship between two nominal (categorical) variables. The
frequency of each category for one nominal variable is compared across the
categories of the second nominal variable. The data can be displayed in a
contingency table where each row represents a category for one variable and
each column represents a category for the other variable.
✓ The Kruskal-Wallis H test (sometimes also called the "one-way ANOVA on
ranks") is a rank-based nonparametric test that can be used to determine if
there are statistically significant differences between two or more groups of an
independent variable on a continuous or ordinal dependent variable. It is
considered the nonparametric alternative to the one-way ANOVA, and an
extension of the Mann-Whitney U test to allow the comparison of more than two
independent groups.
✓ The Kendall rank correlation coefficient or Kendall's tau statistic is used to
estimate a rank-based measure of association. This test may be used if the
data do not necessarily come from a bivariate normal distribution. Tau is
the Kendall correlation coefficient.

118
 ASSESSMENT TASK
Exercise 3.2 - a
Chi Square Test

Name: __________________________ Program & section: __________

Date: ___________________________ Score: ____________________

1. Table below contains information from a survey among 499 participants

classified according to their age groups. The second column shows the
percentage of obese people per age class among the study participants. The
last column comes from a different study at the national level that shows the
corresponding percentages of obese people in the same age classes in the
Philippines. Perform a hypothesis test at the 5% significance level to determine
whether the survey participants are a representative sample of the Philippine
obese population.

Expected Philippine
Age Class (Years) Obese (Percentage)
Average (Percentage)
20–30 75.0 32.6
31–40 26.5 32.6
41–50 13.6 36.6
51–60 21.9 36.6
61–70 21.0 39.7

119
 ASSESSMENT TASK
Exercise 3.2 - b
Sign Test Wilcoxon Matched -Paired Signed-Ranks Test

Name: __________________________ Program & section: __________

Date: ___________________________ Score: ____________________

1. A study is run to evaluate the effectiveness of an exercise program in

reducing systolic blood pressure in patients with pre-hypertension (defined
as a systolic blood pressure between 120-139 mmHg or a diastolic blood
pressure between 80-89 mmHg). A total of 15 patients with pre-
hypertension enroll in the study, and their systolic blood pressures are
measured. Each patient then participates in an exercise training program
where they learn proper techniques and execution of a series of exercises.
Patients are instructed to do the exercise program 3 times per week for 6
weeks. After 6 weeks, systolic blood pressures are again measured. The
data are shown below.
Systolic Blood Pressure
Systolic Blood Pressure
Patient (Before Exercise
(After Exercise Program)
Program)
1 125 118
2 132 134
3 138 130
4 120 124
5 125 105
6 127 130
7 136 130
8 139 132
9 131 123
10 132 128
11 135 126
12 136 140
13 128 135
14 127 126
15 130 132

120
 ASSESSMENT TASK
Exercise 3.2 -c
Mann – Whitney U Test

Name: __________________________ Program & section: __________

Date: ___________________________ Score: ____________________
1. A new approach to prenatal care is proposed for pregnant women living in a rural
community. The new program involves in-home visits during the course of
pregnancy in addition to the usual or regularly scheduled visits. A pilot randomized
trial with 15 pregnant women is designed to evaluate whether women who
participate in the program deliver healthier babies than women receiving usual
care. The outcome is the APGAR score measured 5 minutes after birth. Recall that
APGAR scores range from 0 to 10 with scores of 7 or higher considered normal
(healthy), 4-6 low and 0-3 critically low. The data are shown below.

Usual Care 8 7 6 2 5 8 7 3
New Program 9 9 7 8 10 9 6

Is there statistical evidence of a difference in APGAR scores in women

receiving the new and enhanced versus usual prenatal care?

121
 ASSESSMENT TASK
Exercise 3.2 - d
X2 Test for k Independent Samples

Name: __________________________ Program & section: __________

Date: ___________________________ Score: ____________________

1. Pancreaticoduodenectomy (PD) is a procedure that is associated with

considerable morbidity. A study was recently conducted on 553 patients who
had a successful PD between January 2009 and December 2019 to determine
whether their Surgical Apgar Score (SAS) is related to 30-day perioperative
morbidity and mortality. The table below gives the number of patients
experiencing no, minor, or major morbidity by SAS category.

What would be an appropriate statistical test to examine whether there is an

association between Surgical Apgar Score and patient outcome? Using 14.13
as the value of the test statistic for these data, carry out the appropriate test at
a 5% level of significance.

Surgical Apgar Major morbidity

No morbidity Minor morbidity
Score or mortality
0–4 21 20 16
5–6 135 71 35
7 – 10 158 62 35

122
 ASSESSMENT TASK
Exercise 3.2 -e
Kruskal – Wallis Test One-way ANOVA by Ranks

Name: __________________________ Program & section: __________

Date: ___________________________ Score: ____________________

1. Here is a group of six different rats with their swim speeds on different days
as a result of a change in the water temperature. The question is whether
or not changing the water temperature in the maze affects the swimming
speed.

Animal 20 oC 24 oC 27 oC 32oC
1 39 38 34 31
2 29 27 24 22
3 36 37 29 19
4 25 29 18 24
5 31 27 20 20
6 24 33 30 33

123
 ASSESSMENT TASK
Exercise 3.2 - f
Kendall Rank Correlation Coefficient: r

Name: __________________________ Program & section: __________

Date: ___________________________ Score: ____________________

1. Ten teachers are being ranked for promotion by the University President
and Representative from Faculty Association. The data are presented
below. Determine the degree of relationship between the two.
Representative from Faculty
University President
Association
1 5
2 4
3 10
4 9
5 1
6 2
7 6
8 7
9 8
10 3

124
References:

1. Non – Parametric Test. Retrieved from

https://corporatefinanceinstitute.com/resources/knowledge/other/nonpara
metric-
tests/#:~:text=What%20are%20Nonparametric%20Tests%3F,to%20as%2
0distribution%2Dfree%20tests.
2. Chi-Square goodness of fit test. Retrieved from
https://www.statisticssolutions.com/chi-square-goodness-of-fit-test/
3. Chi-Square goodness of fit test. Retrieved from
https://stats.libretexts.org/Bookshelves/Introductory_Statistics
4. Chi-Square goodness of fit test. Retrieved from
https://www.google.com/search?q=solution+of+ch+square+test&oq=soluti
on+of+ch+square+test&aqs=chrome..69i57j69i60.6543j0j7&sourceid=chro
me&ie=UTF-8
5. Table of Values. Retrieved from
https://onlinepubs.trb.org/onlinepubs/nchrp/cd-22/manual/v2appendixc.pdf
6. Wilcoxon Signed – Rank Test. Retrieved from
https://sphweb.bumc.bu.edu/otlt/mph-
modules/bs/bs704_nonparametric/BS704_Nonparametric6.html
7. Wilcoxon Signed – Rank Test. Retrieved from https://ncss-
wpengine.netdna-ssl.com/wp-
content/themes/ncss/pdf/Procedures/PASS/Wilcoxon_Signed-
Rank_Tests.pdf
8. Wilcoxon Signed – Rank Test. Retrieved from
https://statistics.laerd.com/spss-tutorials/wilcoxon-signed-rank-test-using-
spss-statistics.php
9. Wilcoxon Signed – Rank Test. Retrieved from
https://www.statisticshowto.com/wilcoxon-signed-rank-
test/#:~:text=Use%20the%20smaller%20of%20W,each%20of%20t%20tie
d%20ranks.
10. Friedman Test. Retrieve from https://www.chegg.com/homework-
help/questions-and-answers/friedman-test-complete-hypothesis-test-use-
friedman-test-problem-carefully-read-article-fr-q21765281

125
11. Friedman Test. Retrieve from
https://www.sciencedirect.com/topics/medicine-and-dentistry/friedman-
test#:~:text=The%20Friedman%20test%20is%20an,treatments%20have%
20the%20same%20distributions.&text=Rank%20the%20data%20within%2
0each,treatment%20outcomes%20for%20each%20patient).
12. Kruskal – Wallis Test. Retrieved from https://statistics.laerd.com/spss-
tutorials/kruskal-wallis-h-test-using-spss-statistics.php
13. Chi square Test for Independent k sample. Retrieved from
https://www.statisticssolutions.com/non-parametric-analysis-chi-square/
14. Chi square Test for Independent k sample. Retrieved from
https://sphweb.bumc.bu.edu/otlt/MPH-
Modules/BS/BS704_HypothesisTesting-
ChiSquare/BS704_HypothesisTesting-ChiSquare3.html

15. Kendall Rank Correlation Coefficient: r. Retrieved from

Stephanie Glen. "Kendall’s Tau (Kendall Rank Correlation Coefficient)"
From StatisticsHowTo.com: Elementary Statistics for the rest of
us! https://www.statisticshowto.com/kendalls-tau/
16. Kendall Rank Correlation Coefficient: r. Retrieved from
https://www.statisticssolutions.com/correlation-pearson-kendall-spearman/
17. Kendall Rank Correlation Coefficient: r. Retrieved from
https://www.statisticssolutions.com/correlation-pearson-kendall-spearman/
18. Kendall Rank Correlation Coefficient: r. Retrieved from
https://towardsdatascience.com/kendall-rank-correlation-explained-
dee01d99c535

126