Statistics and Probability

COURSE MATERIAL NO. 7

https://iwannagetphysical.blogspot.com/2019/07/friday-funny-1622-before-and-after.html?m=1

Term 2, AY 2020 – 2021 Statistics and Probability Page | 1

 WHAT WILL YOU
LEARN? HYPOTHESIS
This module has been designed to
help students….

 Determine the
TESTING: TWO
POPULATIONS
procedures in hypothesis
testing involving two
populations.
 Solve problems involving
hypothesis testing of two
means.
 Identify when to use z-
test or t-test in
hypothesis testing
involving two means.

https://www.pinterest.ph/pin/89298005081195023/

Term 2, AY 2020 – 2021 Statistics and Probability Page | 2

 Pre-Activity
WHATS’ INSIDE
Materials: THIS MODULE?
Paper
Pen or Pencil
Calculator
2 Learning Outcomes

Facebook
3 Pre-Activity

Instructions: Testing Two Large

1. Randomly choose 20 of your classmates (10 boys and
4 Independent Samples
10 girls).
2. Check their Facebook account and determine the 10 Exercise 7.1
number of friends they have.
3. Record the gathered data in the table below. Testing Two Small
11 Independent Samples
4. Using the topic in the course material, solve the
problem below. Testing Two Small
5. At 0.05 level of significance, is it enough to claim that 15 Dependent Samples
the number of friends of male Facebook users is less than
the number of friends of female Facebook users? Exercise 7.2
Male Number of Female Number of
20
Fb friends Fb friends
1 1
2 2
22 Worksheet

3 3
4 4 Lesson Summary
5 5 25
6 6
7 7
8 8 26 References

9 9
10 10

Term 2, AY 2020 – 2021 Statistics and Probability Page | 3

 FOCUS Introduction
QUESTIONS
This course material will deal with the tool in making
inferences between two populations. In comparing two
 What are the common populations, we need two samples also, one from each. Two
problems solvable by kinds of samples we discussed from the past course materials
hypothesis testing of are independent and dependent samples.
dependent populations? The independence and dependence of the two sample
 In what problem you will depends on the way that the data are acquired. Also, it
consider the sign of the depends on the study being done.
difference of two
dependent sample? Testing Two Large Independent Samples

We will compare the means of two populations,

say 𝜇1 𝑎𝑛𝑑 𝜇2 by considering their difference.

The inference will be about 𝜇1 − 𝜇2 that will be based in the

test on the observed sample means 𝑋̅1 − 𝑋̅2.

Two Sample Z-test

Assumptions in z-test for Independent Populations

1. Subjects are randomly chosen.
2. The two population are independent to each other.
3. Populations are normally distributed.
4. Populations standard deviation are known.
5. Each population has a size of 30 or above.

Steps for Two Sample Z-test

Step 1: Formulate the null and alternative hypothesis

H 0 : 1  2
H1 : 1  2 or 1  2 or 1  2

The null hypothesis states that there is no significant

difference between the means of the two population while the
alternative hypothesis states that there is a significant
difference between the two means, one is said to be higher or
lower than the other, or simply different from each other.

Term 2, AY 2020 – 2021 Statistics and Probability Page | 4

 Step 2: Determine if the given problem is a two-tailed test or a
one-tailed test.

1  2 : two  tailed test

1  2 : right  tailed test
1  2 : left  tailed test

Step 3: Set the level of significance.

Step 4: Determine the critical value 𝑧𝑐𝑟𝑖𝑡 .

Test 𝛼 = 0.05 𝛼 = 0.025 𝛼 = 0.01

Two-tailed
±1.960 ±2.24 ±2.576
test
Left-tailed
−1.645 −1.960 −2.326
test
Right-tailed
1.645 1.960 2.326
test

Step 5: Solve for the computed value 𝑧𝑐𝑜𝑚𝑝 .

zcomp 
X 1  X 2    1  2 
( Formula 1)
 12  22

n1 n2

zcomp 
X 1  X 2    1  2 
( Formula 2 )
s12 s2 2
 If the population standard
n1 n2
deviation 𝜎 is unknown,
use Formula 2.
where
1 , 2  population means
X 1 , X 2  sample means
 1 ,  2  population standard deviation
s1 , s2  sample standard deviation
n1 , n2  sample size

Term 2, AY 2020 – 2021 Statistics and Probability Page | 5

 Step 6: Compare the critical value and the computed value.

Step 7: Make a statistical decision.

If |𝑧𝑐𝑜𝑚𝑝 | < |𝑧𝑐𝑟𝑖𝑡 | , then do not reject 𝐻0 .

If |𝑧𝑐𝑜𝑚𝑝 | ≥ |𝑧𝑐𝑟𝑖𝑡 |, then reject 𝐻0 .

Step 8: State the conclusion.

Examples:
a. A businessman wants to buy a new property for his next
business. He compares the price of townhouses in Quezon
City and in Makati City and claim that there is a difference in
prices. The gathered data is shown below. Is there enough
evidence to say that there is a difference between the price of
townhouses in Quezon City and in Makati City? Use 𝛼 = 0.05.

Quezon City Makati City

X 1  Php 2,100, 000 X 2  Php 2,320, 000
s1  Php 255, 000 s2  Php 315, 000
n1  46 n2  49

Solution:

Step 1: Formulate the null and alternative hypothesis.

H 0 : 1  2
H1 : 1  2 (claim)

Step 2: Determine if the given problem is a two-tailed test or a

one-tailed test.

Since the symbol in the alternative hypothesis is " ≠ " , then

we use a two-tailed test.

Step 3: Set the level of significance.

𝛼 = 0.05

Term 2, AY 2020 – 2021 Statistics and Probability Page | 6

 Step 4: Determine the critical value 𝑧𝑐𝑟𝑖𝑡 .

𝑧𝑐𝑟𝑖𝑡 = ±1.960

Step 5: Solve for the computed value 𝑧𝑐𝑜𝑚𝑝 . Since the given in the problem is
the standard deviation of the
zcomp 
X 1  X 2    1  2  sample, we use Formula 2.
s12 s2 2

n1 n2

zcomp 
 2,100, 000  2,320, 000    0 
 255, 000    315, 000 
2 2

We are assuming that there is no

46 49 significant difference between the
population means. So,
zcomp 
 220, 000    0 
𝜇1 = 𝜇2
1, 413,586,957  2, 025, 000, 000
𝜇1 − 𝜇2 = 0
220, 000
zcomp 
58, 639.46587

zcomp  3.752

Step 6: Compare the critical value and the computed value.

|𝑧𝑐𝑜𝑚𝑝 | = 3.752 > 1.960 = |𝑧𝑐𝑟𝑖𝑡 |

Step 7: Make a statistical decision.

Reject the null hypothesis.

Term 2, AY 2020 – 2021 Statistics and Probability Page | 7

 Step 8: State the conclusion.

Since we reject the null hypothesis, we accept the alternative

hypothesis.

Hence, there is a significant difference between the prices of

townhouses in Quezon City and Makati City.

b. The Human Resource Department wants to know who

among the male or female employees is more productive. The
male employees wants to prove that they are more
productive. The data below shows the average evaluation of
productiveness of 35 male employees and 42 female
employees.
Male Female
X 1  96 X 2  92
s1  10 s2  6
Since the problem claims that
Is there enough evidence that the male employees is more
the male employees is more
productive than the female employees? Use 𝛼 = 0.01.
productive than the female
employees, we use “>” pertaining
Solution:
that the average of the male
employees productiveness is
Step 1: Formulate the null and alternative hypothesis.
higher than the female’s.

H 0 : 1  2
H1 : 1  2 (claim)

Step 2: Determine if the given problem is a two-tailed test or a

one-tailed test.

Since the symbol in the alternative hypothesis is " > " , then
we use a right-tailed test.

Step 3: Set the level of significance.

𝛼 = 0.01

Step 4: Determine the critical value 𝑧𝑐𝑟𝑖𝑡 .

𝑧𝑐𝑟𝑖𝑡 = 2.326

Term 2, AY 2020 – 2021 Statistics and Probability Page | 8

 Step 5: Solve for the computed value 𝑧𝑐𝑜𝑚𝑝 .

zcomp 
X 1  X 2    1  2 
s12 s2 2

n1 n2

zcomp 
 96  92    0 
10    6 
2 2

35 42

zcomp 
 4   0
2.85714  0.85714

4
zcomp 
1.92725

zcomp  2.075

Step 6: Compare the critical value and the computed value.

|𝑧𝑐𝑜𝑚𝑝 | = 2.075 < 2.326 = |𝑧𝑐𝑟𝑖𝑡 |

Step 7: Make a statistical decision.

Do not reject the null hypothesis.

Step 8: State the conclusion.

Since we do not reject the null hypothesis, we accept the null

hypothesis.

Hence, there is no enough evidence for the claim that the male
employees is more productive than the female employees.

Term 2, AY 2020 – 2021 Statistics and Probability Page | 9

 Exercise 7.1
Two Large Populations

Solve the following.

1. A survey found that the average monthly rental fee of an

ordinary dorm in NCR is Php 5,100 and in Region III is
Php 4,500. The average is obtained from two set pf samples
with 35 dorms in NCR and 32 dorms in Region III with a
standard deviation of Php 200 and Php 230 respectively. At
𝛼 = 0.05, is there a sufficient evidence to say that the dorm
rate in NCR is higher than in Region III?

Term 2, AY 2020 – 2021 Statistics and Probability Page | 10

 Testing Two Small Populations

The hypothesis test used for small populations is called t-

test. It is used for samples with a size less than 30.

Testing the Means between Two Independent Samples

Assumptions in t-test for Independent Samples

1. Subjects are randomly selected.
2. Groups are independent to each other.
3. Populations variances are homogeneous.
4. Populations is normally distributed.

Steps for Two Independent Sample t-test

Step 1: Formulate the null and alternative hypothesis

H 0 : 1  2
H1 : 1  2 or 1  2 or 1  2

Step 2: Determine if the given problem is a two-tailed test or a

one-tailed test.

1  2 : two  tailed test

1  2 : right  tailed test
1  2 : left  tailed test

Step 3: Set the level of significance.

Step 4: Calculate the degrees of freedom (𝑑𝑓 = 𝑛1 + 𝑛2 − 2). Check Course Material 6
Determine the critical value 𝑡𝑐𝑟𝑖𝑡 . for the table for Critical
Values for T-test.

Term 2, AY 2020 – 2021 Statistics and Probability Page | 11

 Step 5: Solve for the computed value 𝑡𝑐𝑜𝑚𝑝 .

tcomp 
X 1  X 2    1  2 
 n1  1 s12   n2  1 s2 2 1 1

n1  n2  2 n1 n2

where
1 , 2  population means
X 1 , X 2  sample means
 1 ,  2  population standard deviation
s1 , s2  sample standard deviation
n1 , n2  sample size

Step 6: Compare the critical value and the computed value.

Step 7: Make a statistical decision.

If |𝑡𝑐𝑜𝑚𝑝 | < |𝑡𝑐𝑟𝑖𝑡 | , then do not reject 𝐻0 .

If |𝑡𝑐𝑜𝑚𝑝 | ≥ |𝑡𝑐𝑟𝑖𝑡 |, then reject 𝐻0 .

Step 8: State the conclusion.

Examples:

a. A researcher wants to determine whether the monthly

salary of SHS teachers in private school and public school
differs. She randomly selects sample of SHS teachers. From
each group, she calculated the means and standard deviation.
The data is shown below. At 𝛼 = 0.05, can she claim that the
monthly salary of SHS public school teacher is higher than
SHS private school teachers?

SHS Public School SHS Private School

X 1  Php 24, 000 X 2  Php 21,300
s1  Php 700 s2  Php 850
n1  12 n2  15

Term 2, AY 2020 – 2021 Statistics and Probability Page | 12

 Solution:

Step 1: Formulate the null and alternative hypothesis

H 0 : 1  2
H1 : 1  2 (claim)

Step 2: Determine if the given problem is a two-tailed test or a

one-tailed test.

right  tailed test

Step 3: Set the level of significance.

  0.05

Step 4: Calculate the degrees of freedom.

𝑑𝑓 = 𝑛1 + 𝑛2 − 2

𝑑𝑓 = 12 + 15 − 2

𝑑𝑓 = 25

Determine the critical value 𝑡𝑐𝑟𝑖𝑡 .

Given:
𝑡𝑐𝑟𝑖𝑡 = 1.708
SHS Public SHS Private
School School
Step 5: Solve for the computed value 𝑡𝑐𝑜𝑚𝑝 . X 1  Php 24, 000 X 2  Php 21,300
s1  Php 700 s2  Php 850
tcomp 
X 1  X 2    1  2 
n1  12 n2  15
 n1  1 s   n2  1 s2
1
2 2
1 1

n1  n2  2 n1 n2

We are assuming that there is no

tcomp 
 24, 000  21,300    0  significant difference between the
population means. So,
12  1 700   15  1850 
2 2
1 1

12  15  2 12 15 𝜇1 = 𝜇2
𝜇1 − 𝜇2 = 0

Term 2, AY 2020 – 2021 Statistics and Probability Page | 13

 tcomp 
 2, 700    0 
11 490, 000   14  722,500  1 1

25 12 15

2700
tcomp 
620200 0.15

2700
tcomp   8.852
305.008

Step 6: Compare the critical value and the computed value.

|𝑡𝑐𝑜𝑚𝑝 | = 8.852 > 1.708 = |𝑡𝑐𝑟𝑖𝑡 |

Step 7: Make a statistical decision.

Reject 𝐻0 .

Step 8: State the conclusion.

Since we reject the null hypothesis, then we accept the

alternative hypothesis.

Hence, we have enough evidence to support the claim that

the monthly salary of SHS public school teacher is higher than
the SHS private school teacher.

Term 2, AY 2020 – 2021 Statistics and Probability Page | 14

 Testing the Means between Two Dependent Samples

Assumptions in t-test for Dependent Samples

1. Subjects are randomly selected.
2. Groups are dependent to each other, in other words, each
component is paired to each other.
3. Populations variances are homogeneous.
4. Populations is normally distributed.

Steps for Two Dependent(Paired) Sample t-test

Step 1: Formulate the null and alternative hypothesis

H 0 : D  0  D  0 means that the average

H1 : D  0 or D  0 or D  0 difference is equal to zero, i.e.
there is no significant difference
Step 2: Determine if the given problem is a two-tailed test or a between the paired samples.
one-tailed test.  D  0 means that there is a
significant difference between
 D  0 : two  tailed test
the paired samples.
 D  0 : right  tailed test
 D  0 : left  tailed test  D  0 means that there is an
increase in the difference
Step 3: Set the level of significance.  D  0 means that there is a
decrease in the difference.
Step 4: Calculate the degrees of freedom (𝑑𝑓 = 𝑛 − 1).
Determine the critical value 𝑡𝑐𝑟𝑖𝑡 .

Step 5: Calculate the paired sample t-test.

D  D
tcomp 
sD
n

Check Course Material 6

 D
2
for the table for Critical
D D 2

n Values for T-test.
D sD 
n n 1

Term 2, AY 2020 – 2021 Statistics and Probability Page | 15

 where
D  difference of each paired sample
D  average difference of each paired sample
sD  standard deviation
n  number of paired sample

Step 6: Compare the critical value and the computed value.

Step 7: Make a statistical decision.

If |𝑡𝑐𝑜𝑚𝑝 | < |𝑡𝑐𝑟𝑖𝑡 | , then do not reject 𝐻0 .

If |𝑡𝑐𝑜𝑚𝑝 | ≥ |𝑡𝑐𝑟𝑖𝑡 |, then reject 𝐻0 .

Step 8: State the conclusion.

Examples:
a. The management of ABC Company designed an incentive
plan for their employees. They randomly choose 9 employees
and record their weekly sales before and after the incentive
plan is implemented.

Employee 1 2 3 4 5
Before 3,000 2,500 4,100 3,700 1,900
After 4,100 2,250 4,200 3,100 2,400

Employee 6 7 8 9
Before 2,500 1,200 2,500 3,000
After 3,200 2,300 1,600 3,000

Was there a significant increase in the weekly sales of the

employees due to the incentive plan of the management? Use
𝛼 = 0.05 significance level.

Term 2, AY 2020 – 2021 Statistics and Probability Page | 16

 Solution:

Step 1: Formulate the null and alternative hypothesis

H 0 : D  0
H1 :  D  0 (claim)

Step 2: Determine if the given problem is a two-tailed test or a

one-tailed test.

right-tailed test

Step 3: Set the level of significance.

𝛼 = 0.05

Step 4: Calculate the degrees of freedom (𝑑𝑓 = 𝑛 − 1).

𝑑𝑓 = 9 − 1 = 8

Determine the critical value 𝑡𝑐𝑟𝑖𝑡 .

𝑡𝑐𝑟𝑖𝑡 = 1.860

Step 5: Calculate the paired sample t-test.

To compute for the test value, we need first to determine

the respective differences 𝐷. We will use a tabular solution to
solve the computed value

Term 2, AY 2020 – 2021 Statistics and Probability Page | 17

 Employee Before 𝑋1 After 𝑋2 𝐷 = 𝑋2 − 𝑋1 𝐷2 = (𝑋2 − 𝑋1 )2
1 3,000 4,100 1,100 1,210,000
2 2,500 2,250 -250 62,500
3 4,100 4,200 100 10,000
4 3,700 3,100 -600 360,000
5 1,900 2,400 500 250,000
6 2,500 3,200 700 490,000
7 1,200 2,300 1,100 1,210,000
8 2,500 1,600 -900 810,000
9 3,000 3,000 0 0
∑ 𝐷 = 1,750 ∑ 𝐷2 = 3,062,500

Then, compute for the mean of the differences.

D
D
n

1, 750
D  194.44
9

Determine the standard deviation of the differences.

 D
2

D 2

n
sD 
n 1

1, 750 
2

3, 062,500 
sD  9
9 1

3, 062,500  340, 777.7778

sD 
8

2, 721, 722.222
sD 
8

sD  583.28

Term 2, AY 2020 – 2021 Statistics and Probability Page | 18

 Solve for the computed value.

D  D
tcomp 
sD
n

194.44  0
tcomp 
583.28
9

194.44
tcomp 
194.427

tcomp  1.00

Step 6: Compare the critical value and the computed value.

|𝑡𝑐𝑜𝑚𝑝 | = 1.00 < 1.860 = |𝑡𝑐𝑟𝑖𝑡 |

Step 7: Make a statistical decision.

Do not reject null hypothesis.

Step 8: State the conclusion.

Since we do not reject the null hypothesis, we accept the null

hypothesis. Hence, there is no enough evidence that there is
an increase in sale after the implementation of the incentive
program.

Term 2, AY 2020 – 2021 Statistics and Probability Page | 19

 Exercise 7.2
Two Small Populations

Solve the following problems.

1. Given a sample of 𝑛1 = 21 from a population with 𝑠1 =
11.25, and another sample of 𝑛2 = 15 from a different
population with 𝑠2 = 7.6. If 𝛼 = 0.05, what is the statistical
̅̅̅1 = 91.2 , 𝑋
decision given that 𝑋 ̅̅̅2 = 85.2, and the hypotheses
are;

H 0 : 1  2
H1 : 1  2

Term 2, AY 2020 – 2021 Statistics and Probability Page | 20

 2. The principal wants to determine if the teaching strategies
of his teachers used in online teaching is effective or not. A
sample of 10 students is randomly chosen. They took a pre-
test and a post-test. The scores of each student is listed below.

Student 1 2 3 4 5
Pre-test 35 25 27 34 15
Post-test 38 27 50 35 26

Student 6 7 8 9 10
Pre-test 21 18 37 25 35
Post-test 27 17 24 30 25

Use 𝛼 = 0.01. Is there enough evidence that the teaching

strategies are effective?

Term 2, AY 2020 – 2021 Statistics and Probability Page | 21

 WORKSHEET

I. Determine if the given problem is solvable by Z-test or T-test. Also, identify if the given problem
samples are independent or dependent. Then, solve the problem.

1. A researcher wants to compare the hourly rates of delivery drivers in two major cities in Metro
Manila. He randomly selects drivers from both cities and records their hourly rates (in pesos). It is
shown below:

Quezon
350 315 275 330 300 270 255 315 310
City
Makati
400 275 315 350 380 410 365 280 340
City

Is there sufficient evidence to say that the hourly rates between the two cities are not the same, at 𝛼 =
0.05?

Term 2, AY 2020 – 2021 Statistics and Probability Page | 22

 2. A manager of a business firm randomly selected 15 of her employees to attend a seminar-workshop
as an aid to their productiveness in the workplace. The table below shows the workload done per
week of the 15 employees before and after the seminar-workshop. At 0.01 level of significance, is
there a sufficient evidence to conclude that the seminar-workshop increase the performance level of
the employees?

Employee 1 2 3 4 5 6 7 8
Before 5 8 11 12 5 7 8 9
After 7 11 7 9 14 11 12 9

Employee 9 10 11 12 13 14 15
Before 11 12 6 7 9 12 13
After 10 13 9 11 9 13 15

Term 2, AY 2020 – 2021 Statistics and Probability Page | 23

 3. A survey found that the average bed space room rate in Caloocan City is Php 2,500 and the average
rate in Pasig City is Php 3,100. The standard deviations were Php 310, and Php 290, respectively.
Assume that the data were obtained from two samples of 40 rooms from Caloocan City and 47 rooms
from Pasig City, is there enough evidence to say that the average rate of Caloocan City is less than the
average rate in Pasig City. Use 𝛼 = 0.05.

Term 2, AY 2020 – 2021 Statistics and Probability Page | 24

 LESSON SUMMARY

Two samples are said to be dependent if one is related to another.

Dependent samples are also known as paired samples because each element of
one sample is paired to another element on the other sample. Usually the
problems involving a before-after situations is solvable by hypothesis testing of
dependent samples. It can also solve comparison between two objects which are
related to each other such as two different academic subjects, etc.

If a give problem shows a before-after relationship, signs are needed and

useful. A positive sign means an increase from the before value to after value. On
the other hand, a negative sign indicates a decrease in value. If you are only
concern with the difference of the two elements regardless of the direction
(increase or decrease), then you take the absolute value of the difference.

Term 2, AY 2020 – 2021 Statistics and Probability Page | 25

 REFERENCES

 Anderson, D. R., et al. 2017. Statistics for

Business and Economics, 13th Edition.
Cengage Learning.
 Bonghanoy, G. et al. 2017. Statistics and
Probabilty for Senior High School.
Mutya Publishing
 Gonzales, J. & Nocon, R. 2015. Essential
Statistics. MaxCor Publishing House, Inc.
272pp.
 Montero-Galliguez, T. et al. 2016.
Fundamentals of Statistical Analysis. C &
E Publishing, Inc. 245pp.
 Sirug, W.S. 2011. Basic Probabilities and
Statistics. Mindshapers Co, Inc. 343pp.

Contributors:
Aserit, Pia Lorraine O.
Ignacio, Rinalyn L.
Rivera, Rey-Ann B.

Term 2, AY 2020 – 2021 Statistics and Probability Page | 26