Statistics for Business and Economics (14e)

Tests we study as a part of our course

1. Chi-Square test
2. Runs test
3. Mann–Whitney U test
4. Wilcoxon matched-pairs signed rank test
5. Kruskal–Wallis test
6. Friedman test
7. Spearman’s rank correlation

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 1
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

2. Runs Test
✓The Runs Test is a non-parametric test used to determine whether a sequence of
data points is random. It checks whether there is any pattern or systematic
arrangement in the data.
✓A run is defined as a sequence of similar symbols (e.g., positive and negative signs, or
heads and tails) that is followed or preceded by a different symbol or no symbol at
all.
Applications of Runs Test:
• To test the randomness of a sequence.
• In quality control to check the randomness of defects.
• For binary or continuous data where randomness is assumed.

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 2
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Fundamental Principles of the Run Test

Reject randomness if the number of runs is very low or very high.
Example: The sequence of genders FFFFFMMMMM is not random
because it has only 2 runs, so the number of runs is very low.

Example: The sequence of genders FMFMFMFMFM is not random

because there are 10 runs, which is very high. It is important to note
that the runs test for randomness is based on the order in which the
data. It is not based on the frequency of the data.

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 3
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Formula for Runs Test:

• Mean of the number of runs (μR​): Where:
•R: Observed number of
runs.

• Standard deviation of the number of runs •n1​: Number of first

type of data).

•n2​: Number of second

type of data).

• Z-score:

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 4
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

A sample problem on Runs test

A sequence of defective (D) and non-defective (N) items is recorded as follows:
D,N,D,N,N,D,D,N,N,D,N,N,D,D,N,N
Check if the sequence is random at a 5% significance level.
•Null Hypothesis (H0):
The sequence of defective (D) and non-defective (N) items is random.
•Alternative Hypothesis (H1​):
The sequence of defective (D) and non-defective (N) items is not random.

Step 1: Count n1​ and n2​:

• n1​: Number of D = 7 (defective items).
• n2​: Number of N = 9 (non-defective items).
© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 5
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Step 2: Count observed number of runs (RRR):

Break the sequence into runs: 16 runs

Step 3: Calculate the expected number of runs (μR​):

Step 4: Calculate the standard deviation of the runs:

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 6
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Step 6: Compare Z-score with critical value:

At a 5% significance level (two-tailed), the critical Z-values are ±1.96.

Since Z=1.64 which lies within the range of −1.96 too +1.96, we fail to reject the null
hypothesis.

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 7
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Second sample problem on Runs test

A quality control engineer recorded the sequence of acceptable (A)
and defective (D) items in a production process as follows:
• A, A, D, A, D, D, A, A, D, A, A, A, D, D, A, D, A, D, D, A
Determine whether the sequence is random at a 5% significance
level.

we fail to reject the null

hypothesis.

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 8
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Third sample problem on Runs test

• A marketing team recorded the daily sales performance of a

product as either high (H) or low (L) for 20 days. The
sequence of sales performance is as follows:
H, L, L, H, H, L, H, L, L, L, H, H, L, L, H, H, H, L, H, L
• Perform a Runs Test for Randomness at the 5% significance
level to determine whether the sales performance sequence
is random.

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 9
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

3. Mann-Whitney U Test Overview

The Mann-Whitney U test is a non-parametric test used to determine
whether there is a significant difference between the distributions of two
independent groups. It is often used as an alternative to the independent
samples t-test when the assumptions of normality or equal variances are
not met.

Uses of Mann-Whitney U Test

▪ Compare two independent groups on a continuous or ordinal outcome.
▪ Analyze small sample sizes when the normality assumption cannot be satisfied.
▪ Assess whether one group tends to have higher or lower values than the other.

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 10
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Formula for Mann-Whitney U test

• For two independent groups, A and B, with sample sizes n1 and n2​:
1. Compute R1​: The sum of ranks for Group A.
2. Compute R2​: The sum of ranks for Group B.

The Mann-Whitney U-statistic for each group is:

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 11
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Formula for Mann-Whitney U test

• Z-Score for Mann-Whitney U Test
For larger sample sizes, a Z-score approximation can be used:

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 12
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Sample Problem
Two groups of students were tested on their math skills. The scores are as follows:
Group A (n1=5): 8,12,15,20,25
Group B (n2​=6): 5,10,10,18,22,30
Determine if there is a significant difference between the two groups using the
Mann-Whitney U test at a 5% significance level.

Null Hypothesis (𝐻0​ ):The distributions of the two groups (Group A and Group B) are
identical. There is no significant difference between the scores of the two groups.

Alternative Hypothesis (𝐻1):The distributions of the two groups are not identical. There is
a significant difference between the scores of the two groups.

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 13
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

solution For the Mann-Whitney U test, the ranking of

variables can be done in either ascending or
• Step 1: Rank the combined data descending order
Combine and rank all the scores:
5,8,10,10,12,15,18,20,22,25,
Ranks are: • Assign ranks to each group:
1,2,3.5,3.5,5,6,7,8,9,10,11 • Group A: 8(2),12(5),15(6),20(8),25(10)
• Group B: 5(1),10(3.5,3.5),18(7),22(9),30(11)

Step 2: Calculate the sum of ranks (R1, R2​)

R1=2+5+6+8+10=31
R2=1+3.5+3.5+7+9+11=35

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 14
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Step 3: Calculate U1 and U2

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 15
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Step 6: Compare Z-score with critical value

At a 5% significance level (two-tailed), the critical Z-values are ±1.96. Since Z=−0.18,
which lies within −1.96, +1.96, we fail to reject the null hypothesis.

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 16
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Practice problem on Mann-Whitney U test

• Problem: Comparing Customer Satisfaction Scores between Two Stores
A retail chain wants to compare customer satisfaction levels between two of its
stores (Store A and Store B). A random sample of customers from each store is
surveyed, and they provide satisfaction ratings on a scale of 1 to 10. The ratings
are as follows:
• Store A (n1=8): 8,7,9,4,1,10,9
• Store B (n2​=7): 6,5,7,3,8,2
• Using a significance level of 5%, determine if there is a significant difference in
customer satisfaction levels between the two stores.

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 17
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

4. Spearman’s Rank Correlation (1 of 2)

• The Pearson correlation coefficient, 𝑟, is a measure of the linear association between two
variables for which interval or ratio data are available.

• The Spearman rank-correlation coefficient, 𝑟𝑠 , is a measure of association between two

variables when only ordinal data are available.

• Values of 𝑟𝑠 can range from –1 to +1, where

• values near 1 indicate a strong positive association between the rankings, and

• values near – 1 indicate a strong negative association between the rankings.

• Values near 0 indicates no correlation

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 18
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

When to use Spearman’s rank correlation test?

•A monotonic relationship means as one variable increases, the other either
consistently increases or consistently decreases, but not necessarily at a
constant rate.
•Input: Uses the ranks of the data, not the raw values.
•Non-parametric: Does not assume a normal distribution of the data.

Suitable when:
•Data is ordinal or ranks are used.
•The relationship between variables is not linear.
•The data contains outliers.

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 19
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Rank Correlation (2 of 2) for Spearman’s rank correlation test

whether ranking of variables should be
Spearman Rank-Correlation Coefficient, 𝑟𝑠 ascending or in descending order or can
we use any one of the ranking order

where:
𝑛 = number of observations being ranked of any one group
𝑥𝑖 = rank of observation 𝑖 with respect to the first variable
𝑦𝑖 = rank of observation 𝑖 with respect to the second variable
𝑑𝑖 = 𝑥𝑖 − 𝑦𝑖

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 20
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Test for Significant Rank Correlation

We may want to use sample results to make an inference about the
population rank correlation ps.
• To do so, we must test the hypotheses:

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 21
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Rank Correlation (1 of 7)
Sampling Distribution of 𝑟𝑠 when 𝑝𝑠 = 0

• Mean

• Standard Deviation

• Distribution Form, test statistic

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 22
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Rank Correlation (2 of 7)
Example: Crennor Investors
Crennor Investors provides a portfolio management service for its clients. Two of Crennor’s
analysts ranked ten investments as shown on the next slide. Use rank correlation, with  =
0.10, to comment on the agreement of the two analysts’ rankings.

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 23
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Rank Correlation (3 of 7)
Example: Crennor Investors
• Analysts’ Rankings are shown in the table.
• Hypotheses

Investment A B C D E F G H I J
Analyst 1 1 4 9 8 6 3 5 7 2 10
Analyst 2 1 5 6 2 9 7 3 10 4 8

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 24
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Rank Correlation (4 of 7)
Investment Analyst #1 Ranking Analyst #2 Ranking Differ. Left parenthesis
Differ. right
parenthesis squared

A 1 1 0 0

B 4 5 -1 1

C 9 6 3 9

D 8 2 6 36

E 6 9 -3 9

F 3 7 -4 16

G 5 3 2 4

H 7 10 -3 9

I 2 4 -2 4

J 10 8 2 4

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 25
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Rank Correlation (5 of 7)
Sampling Distribution of 𝑟𝑠 Assuming No Rank Correlation

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 26
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Rank Correlation (6 of 7)

• Rejection Rule

• Test Statistic

• 𝑝-value

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 27
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Rank Correlation (7 of 7)
Conclusion
Do no reject 𝐻0 . The 𝑝-value > . There is not a significant rank correlation. The two
analysts are not showing agreement in their ranking of the risk associated with the
different investments.

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 28
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Sample Problem on Spearman’s rank correlation -1

A researcher wants to examine the relationship between exercise frequency (hours per
week) and mental health scores (on a scale of 1 to 50). The data for 7 individuals is given
below: Calculate the Spearman’s Rank Correlation Coefficient (rs).

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 29
password-protected website or school-approved learning management system for classroom use.
 Statistics for Business and Economics (14e)

Sample Problem on Spearman’s rank correlation -2

with repetition of data points
A researcher wants to study the relationship between students' attendance percentage and
their test scores. The data for 6 students is as follows, with some repeated values in both
variables: Calculate the Spearman’s Rank Correlation Coefficient (rs).

© 2020 Cengage. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license distributed with a certain product or service or otherwise on a 30
password-protected website or school-approved learning management system for classroom use.