BUSINESS ANALYTICS
ARJAY T. ALTOVAR
CAS-MNSD INSTRUCTOR
Topic Outline
Basic Statistical Analysis
1. Statistical Hypotheses
2. z-Test and t-Test for One Population
3. z-Test and t-Test for Two Populations
4. Chi-Square Test of Independence
5. Correlation and Regression
6. Basic Statistical Analysis in Excel and JASP
Two Sample Tests
(z-test)
Two Samples Tests
RIGHT-TAILED TEST:
𝐻0 : 𝜇1 = 𝜇2 or 𝐻0 : 𝜇1 − 𝜇2 =0
𝐻1 : 𝜇1 > 𝜇2 𝐻1 : 𝜇1 − 𝜇2 >0
LEFT-TAILED TEST:
𝐻0 : 𝜇1 = 𝜇2 or 𝐻0 : 𝜇1 − 𝜇2 =0
𝐻1 : 𝜇1 < 𝜇2 𝐻1 : 𝜇1 − 𝜇2 <0
TWO-TAILED TEST:
𝐻0 : 𝜇1 = 𝜇2 or 𝐻0 : 𝜇1 − 𝜇2 =0
𝐻1 : 𝜇1 ≠ 𝜇2 𝐻1 : 𝜇1 − 𝜇2 ≠0
Two Samples Tests
z-Test for Two Populations
Assumptions:
1. Both samples are random samples.
2. The samples must be independent of each other. That
is, there can be no relationship between the subjects in
each sample.
3. The standard deviations of both populations must be
known, and if the sample sizes are less than 30, the
populations must be normally or approximately
normally distributed.
Two Samples Tests
When 𝜎1 2 = 𝜎2 2
(x̄1 −x̄2 ) − (𝜇1 − 𝜇2 )
𝑧=
𝜎1 2 𝜎2 2
+
𝑛1 𝑛2
where,
(x̄1 −x̄2 ) is the observed difference
(𝜇1 − 𝜇2 ) is the expected difference
𝜎1 2 𝜎1 2
+ is the standard error of the difference
𝑛1 𝑛2
Two Sample Tests
Z-test
Illustration:
A company wanted to compare the performance of its call center
employees in two different centers located in two different parts of the
country – Philippines, and Malaysia, in terms of the number of tickets
resolved in a day (hypothetically speaking). The company randomly
selected 30 employees from the call center in Philippines and 30
employees from the call center in Malaysia. The company wants to
determine if the performance of the employees in Philippines is different
from the performance of the employees in the Malaysia center. The
following data was collected:
Philippines: x̄1 = 750, 𝑠1 2 = 400
Malaysia: x̄2 = 780, 𝑠2 2 = 625
One Sample Tests
Example : Z-test
A company wanted to compare the performance of its call center employees in two
different centers located in two different parts of the country – Philippines, and Malaysia, in terms
of the number of tickets resolved in a day (hypothetically speaking). The company randomly
selected 30 employees from the call center in Philippines and 30 employees from the call center in
Malaysia. The company wants to determine if the performance of the employees in Philippines is
different from the performance of the employees in the Malaysia center. The following data was
collected:
Philippines: x̄1 = 750, 𝜎1 2 = 400
Malaysia: x̄2 = 780, 𝜎2 2 = 625
STEP 1:
STEP 2:
STEP 3:
STEP 4:
STEP 5:
Two Sample Tests
(t-test)
Two Sample Tests
(t-test)
Welch’s t-test: Variances are unequal
Student’s t-test: Variance are equal
Two Samples Tests
t-Test for Two Populations
1. The samples are random samples.
2. The sample data are independent of one another.
3. When the sample sizes are less than 30, the
populations must be normally or approximately
normally distributed.
*Population standard deviations were not known and
the variables were normally or approximately
normally distributed, or when both sample sizes were
greater than or equal to 30.
Two Samples Tests
Welch t-test when 𝜎1 2 = 𝜎2 2
(x̄1 −x̄2 ) − (𝜇1 − 𝜇2 )
𝑧=
𝑠12 𝑠22
+
𝑛1 𝑛2
where,
𝑠12 𝑠22
+ is the standard error of the difference
𝑛1 𝑛2
With degrees of freedom (Welch–Satterthwaite Approximation)
2
𝑠12 𝑠22
𝑛1 + 𝑛2
𝑑𝑓 =
2 2
𝑠12 𝑠22
𝑛1 𝑛2
+
𝑛1 − 1 𝑛2 − 1
Two Sample Tests
Welch’s t-test
Illustration:
A business analyst wants to investigate whether employees working in
high-stress departments have higher daily absenteeism compared to employees
in low-stress departments. To do this, a sample of employees from high-stress
departments and a sample from low-stress departments were collected, and their
daily absenteeism rates were recorded. Preliminary analysis indicates that the
two samples are normally distributed, but their variances are significantly
different. At a significance level of 0.1, determine whether employees in high-
stress departments have significantly higher absenteeism than those in low-
stress departments. This analysis will help the company assess the impact of
work environment stress on employee attendance and guide potential
interventions.
n ഥ
𝑿 𝒔𝟐
Low Stress 7 3.14 3.7
High Stress 8 11.6 3.9
Two Sample Tests
Example : t-test
The extend to which an infant’s health is affected by parental smoking is an important public
health concern. The following data are the urinary concentrations of cotinine (a metabolite of
nicotine); measurements were taken both from a sample of infants who had been exposed to
household smoke and from a sample of unexposed infants. Assume that both samples are
normally distributed and 𝜎1 2 and 𝜎2 2 are significantly equal. At significance level of 0.1,
determine if there is significant difference between two samples.
STEP 1:
STEP 2:
STEP 3:
STEP 4:
STEP 5:
Two Sample Tests
(t-test)
Welch’s t-test: Variances are unequal
Student’s t-test: Variance are equal
Two Samples Tests
Student’s t-test when 𝜎1 2 = 𝜎2 2
(x̄1 −x̄2 ) − (𝜇1 − 𝜇2 )
𝑡=
(𝑛1 −1)𝑠1 2 +(𝑛2 − 1)𝑠2 2 1 1
+
𝑛1 + 𝑛2 − 2 𝑛1 𝑛2
With degrees of freedom:
𝑛1 + 𝑛2 − 2
Where the pooled variance is,
(𝑛1 −1)𝑠1 2 +(𝑛2 − 1)𝑠2 2
𝑛1 + 𝑛2 − 2
Two Sample Tests
Student’s t-test
Activity 1: (Short Bond Paper)
A business analyst wants to compare the average store size
of retail outlets operated by a company in Region A and Region B.
Preliminary data show that the average store size in Region A is 191
square meters, while the average store size in Region B is 199
square meters. Data were collected from two samples of stores: 8
stores from Region A and 10 stores from Region B, with standard
deviations of 37 and 36 square meters, respectively. Assuming the
store sizes are normally distributed and that the population
variances are equal, the analyst wants to determine, at a
significance level of α = 0.05, whether there is a significant
difference in average store sizes between the two regions.
Two Sample Tests
(dependent /paired samples t-test)
Two Samples Tests
t-Test for Two Populations
Assumptions
1. The sample or samples are random.
2. The sample data are dependent.
3. When the sample size or sample sizes are less than
30, the population or populations must be normally
or approximately normally distributed.
Two Samples Tests
RIGHT-TAILED TEST:
H0: 𝜇𝐷 ≤ 0
H1: 𝜇𝐷 > 0
LEFT-TAILED TEST:
H0: 𝜇𝐷 ≥ 0
H1: 𝜇𝐷 < 0
TWO-TAILED TEST:
H0: 𝜇𝐷 = 0
H1: 𝜇𝐷 ≠ 0
Two Samples Tests
Paired t-test: when samples are
dependent
ഥ −𝜇𝐷
D
𝑡= 𝑠𝐷
ൗ 𝑛
where, degrees of freedom is n-1
∑𝐷 𝑛∑𝐷2 −(∑𝐷)2
ഥ=
D 𝑎𝑛𝑑 𝑠𝐷 =
𝑛 𝑛(𝑛−1)
Two Sample Tests
Paired/ dependent samples t-test
Example :
A business analyst wants to determine whether a new employee
wellness program has an impact on employee productivity. Six employees
were measured for their average daily completed tasks before participating
in the program and again after completing a 6-week wellness program. The
analyst assumes that the productivity scores are approximately normally
distributed. Using a significance level of α = 0.10, the analyst aims to
determine whether there is enough evidence to conclude that the wellness
program has significantly changed employee productivity. This scenario
calls for a paired t-test since the same employees are measured before and
after the program.
Employees 1 2 3 4 5 6
Before (𝑿𝟏 ) 210 235 208 190 172 244
After (𝑿𝟐 ) 190 170 210 188 173 228
One Sample Tests
Example : paired/dependent samples t-test
A dietitian wishes to see if a person’s cholesterol level will change if the diet is
supplemented by a certain mineral. Six subjects were pretested, and then they took the
mineral supplement for a 6-week period. The results are shown in the table. (Cholesterol
level is measured in milligrams per deciliter.) Can it be concluded that the cholesterol
level has been changed at 𝛼=0.10? Assume the variable is approximately normally
distributed.
STEP 1:
STEP 2:
STEP 3:
STEP 4:
STEP 5:
THANK YOU!
ARJAY T. ALTOVAR
CAS-MNSD INSTRUCTOR