Republic of the Philippines

Department of Education
MIMAROPA Region
SCHOOLS DIVISION OF ORIENTAL MINDORO
SAN MARIANO NATIONAL HIGH SCHOOL
San Mariano. Roxas, Oriental Mindoro

Statistics and Probability

Quarter 4 – Week 5

Self-Learning Module 5
HYPOTHESIS TESTING ON
POPULATION PROPORTION

What I Need to Know

 Advertisements make use of population proportion to promote their
products. They would often endorse that 8 out of 10 people has switched to
their product over their competitors or 3 out of 4 cool people use this product.
However, they should be able to test their hypothesis so that they wouldn’t be
accused of false advertisement.
In order for us to fully grasp the concept of hypothesis testing, we must
practice on a lot of problems. With enough practice, the steps in hypothesis
testing will be second nature to us.
After going through this module, you are expected to:
1. Computes for the t-statistic value (population proportion).
(M11/12SP-IVf-1)
2. Draws conclusion about the population proportion based on the test-
statistic value and the rejection region. (M11/12SP-IVf-2)
3. Solves problems involving test hypothesis on the population
proportion. (M11/12SP-IVf-g-1)

What I Know

Determine whether the statement is True or False by checking the appropriate

box.
True False
1. Percentages can be expressed as proportions.
2. A proportion is obtained when a frequency of desired
events is multiplied by the sum of events

If the confidence level is 95%, then 𝜇 𝑖𝑠 0.025.

3. If n=25, the Central Limit Theorem applies.
4.
2
5. When x =124 and n=260, then 𝑋 = 0.48.
�
�
6. The p-value of z=2 is 0.4772.
7. The p-value of z≤-2 is 0.0228.
8. When H0 is rejected, it means that a significant
difference does not exist.
9. When the evidence is not enough, do not reject the null
hypothesis.

1
 10. When the evidence is sufficient to reject the null hypothesis,
a significant difference exists.

Lesson
7
Hypothesis Testing on Population Proportion

In interpreting the hypothesis of a population proportion, the test statistic

is used to interpret whether a data is accepted or rejected. The test-statistic
must lie on a certain region for it to be accepted, much like the die has to give
a result of 3 or 4 for it to be counted. Otherwise, it will be rejected.

What’s In

Let us recall the steps in hypothesis testing using the five-step procedure.
1. Formulate the null (Ho) and alternative (Ha) hypothesis.
2. Set the level of significance. Decide on the test statistic to be used and find
the critical value.
3. Compute the test value, using the test statistic or formula for the test.
4. Make decision whether to accept or reject the null hypothesis.
5. Make a conclusion.

What is It

We have mentioned in the previous lesson that the population mean can
be estimated only for large sample size (n ≥ 30). The same is true in testing a
claim or hypothesis about the population proportion (po).
A population proportion is a fraction of the population that has a
certain characteristic. For example, a researcher who is studying on the rapid
growth of the population wants to determine the proportion of female rats in a
certain region, then he doesn’t need to catch every rat he sees and record its
gender. He only needs a sufficient sample from which he will make inference
about the proportion of female rats.
In the example above, the researcher may initially believe that 50% of
the rat population are female. Out of 50 rats he collected, 23 are female.
2
 To test a claim about population proportion, we use the z-test for
population proportion.
p− p 0

√
z= p0 (1−p 0)
n
where: p = sample proportion
po = given population proportion
n = the sample size
x
p=
n
x = the number of successes
Decision rule:
The decision rule can be one of the following possibilities:
1. Reject the null hypothesis (𝐻𝑜) if 𝑍c < −𝑍t. Otherwise, do not reject
𝐻𝑜.
Reject the null hypothesis (𝐻𝑜) if 𝑍c > 𝑍t. Otherwise, do not
reject 𝐻𝑜.
2.

3. Reject the null hypothesis (𝐻𝑜) if |𝑍c| > 𝑍t. Otherwise, do not reject
𝐻𝑜.

Example 1: Compute the z for each given claim (p o), observed proportion (𝑝)
and the sample size (n).
a.) 𝑝o = 0.3, 𝑝 = 0.4,
b.) 𝑝o = 0.8, 𝑝 = 0.72,
n = 60

c.) 𝑝o = 0.66, 𝑝 = 0.61,

n = 30
n = 40
Solution:
p− p 0 0.4−0.3 0.1 0.1
0.1
a.) z =

0.1
√
p0 (1−p 0) =
n √ 0.3(1−0.3) =
60 √ 0.3(0.7) =
60 √ 0.21 =
60
√ 0.0035

= = 1.69
0.059

3
 p− p 0 0.72−0.8 −0.08 −0.08
−0.08
b.) z =

−0.08
√ n 30√
p0 (1−p 0) = 0.8(1−0.8) =
√ 30 √
0.8(0.2) = 0.16 =
30
√ 0.0053
=

= ̵ 1.10
0.073
p− p 0 0.61−0.66 −0.05
−0.05 −0.05

− 0.67
√ n 40√
c.) z = p0 (1−p 0) = 0.66 (0.34) = 0.2244 =
40 √ =
√ 0.00561 0.0749
=

Example 2. From the above example the researcher wants to test his belief that
50% or 0.50 of the population of rats is female. From the

the claim? Use 𝛼 = 0.05.

collected samples, 23 out of 50 are female. Would this support

Solution: Claim. The manufacturer claims that the average lifetime of his
lightbulbs is 3 years or 36 months
Step 1. State the hypotheses
H0 : 𝑝o = 0.50
Hα : 𝑝o ≠ 0.50
Step 2. Set the level of significance. Determine the critical values
and rejection region.
Since α = 0.05, and Ha used ≠ symbol therefore it is two-
tailed, the critical values are zt = ±1.96 (see table).

Test statistic: z-test

𝜶
Confidence Critical Values Critical Value Critical Value
Level (2- Tailed) (1-Tailed Left) (1-Tailed Right)
90% z = ±1.645 z = −𝟏. 𝟐𝟖 z = 𝟏. 𝟐𝟖 0.10
95% z = ±1.96 z = −𝟏. 𝟔𝟒𝟓 z = 𝟏. 𝟔𝟒𝟓 0.05
99% z = ±2.575 z = −𝟐. 𝟑𝟑 z = 𝟐. 𝟑𝟑 0.01

Step 3. Compute the test-statistic

4
 Given: 𝑝o = 0.50, 𝑝 =
x 23
n = 50, x = 23 = =
n 50
0.46
Solution:
p− p 0 0.46−0.50 −0.04
−0.04
z =
√ p0 (1−p 0) =
n √ 50 √
0.50(0.50) = 0.25 =
50
√ 0.005
=

= − 0.57
−0.04
0.07
Step 4. Decision rule: Reject the null hypothesis (𝐻𝑜) if |𝑍c| >
𝑍t for a two-tailed test. (Negative sign
could be disregarded since the test is two-
tailed.)
Since |-0.57| < 1.96, do not reject H0.

Step 5. Conclusion: There is no sufficient evidence to deny the

researcher’s claim. Thus, 50% of the rat population are
female.

Example 3. Mr. Caberto asserts that fewer than 5% of the bulbs that he sells
are defective. Suppose 300 bulbs are randomly selected, each are
tested, and 10 defective bulbs are found. Does this provide
sufficient evidence for Mr. Caberto to conclude that the fraction
of defective bulbs is less than 0.05? Use α = 0.01.
Solution: Claim: Mr. Caberto asserts that fewer than 5% of the bulbs that he
sells are defective.
Step 1. State the hypotheses
Ho : 𝑝o ≥ 0.05
Ha : 𝑝o < 0.05
Step 2. Set the level of significance. Determine the critical values
and rejection region.
Since α = 0.01, and Ha used < symbol therefore it is one-
tailed left, the critical value is zt = -2.33(see table above).

5
 Step 3. Compute the test-statistic

Given: 𝑝o = 0.05, n = 300, x = 10 𝑝 =

x 10
= =
n 300
0.03
Solution:
p− p 0 0.033−0.05 −0.017
−0.017

−0.017
z=
√ p0 (1−p 0) =
n √ 300 √
0.05(0.95) = 0.0475 =
300
√ 0.000158
=

= −1.35
0.0126

Step 4. Decision rule: Reject the null hypothesis (𝐻𝑜) if 𝑍c <

−𝑍t. Otherwise, do not reject 𝐻𝑜.
Since -1.35 > -2.33, reject H0.
Step 5. Conclusion: The data provides sufficient evidence to Mr.
Caberto’s assertion that fewer than 5% of the bulbs that he
sells are defective.

Example 4. A manufacturing company claimed that more than 55% of the

city’s consumers uses their toothpaste (x). In a random sample of

is it safe to say that the company’s claim was valid? Use 𝛼 = 0.05.
200 consumers, 125 are users of their product. Based on the data,

Solution: Claim: The company claimed that more than 55% of the city’s
consumers uses their toothpaste.
Step 1. State the hypotheses
Ho : 𝑝o ≤ 0.05
Ha : 𝑝o > 0.05
Step 2. Set the level of significance. Determine the critical values
and rejection region.
Since α = 0.05, and Ha used > symbol therefore it is one-
tailed right, the critical value is zt = 1.645 (see table above).
Step 3. Compute the test-statistic

6
 Given: 𝑝o = 0.55, n = 200, x = 125 𝑝 =
x 125
= =
n 200
0.625
Solution:
p− p 0 0.625−0.55 0.075
0.075

0.075
z=
√ p0 (1−p 0) =
n √ 200 √
0.55(0.45) = 0.2475 =
200
√ 0.001238
=

= 2.14
0.035

Step 4. Decision rule: Reject the null hypothesis (𝐻𝑜) if 𝑍c >

𝑍t. Otherwise, do not reject 𝐻𝑜.
Since 2.14 > 1.645, reject H0.
Step 5. Conclusion: The data provides sufficient evidence to company’s claim
that more than 55% of the city’s consumers uses their
toothpaste.

What’s More

Directions: Complete the solution in each of the following problem. Show

your complete solution.
1. In 2012, 1,500 randomly selected pine trees in Baguio were tested for
traces of Bark Beetle infection. It was found that 153 of the trees showed
such traces. Test the hypothesis that more than 10% of the trees have
been infected. (Use 0.05 level of significance)
Solution :
Step 1. Ho : ________
Ha : ________
Step 2. α = _____
zt = ______
7
 Step 3. zc = _______

Step 4. Decision Rule : _________________________

Step 5. Conclusion: ____________________________

What I Have Learned

Directions: Reflect the learning that you gained after taking up this lesson on
“Hypothesis Tests for a Population Proportion) by completing the given
statements below. Do this on your activity notebook. Do not write anything on
this module.
What were your thoughts or ideas about the topic before taking up the lesson?
I thought that___________________________________________________
___________________________________________________
.
What new or additional ideas have you had after taking up this lesson?
I learned that___________________________________________________
___________________________________________________

How are you going to apply your learning from this lesson?
I will apply___________________________________________________
___________________________________________________

References

Books:

Alferez, Merle S. et al. 2006. “Hypothesis Testing. In Statistics and

Probability. 155-158. Imelda Avenue, Cainta. MSA Publishing House.

8
Belecina, R.R.; Baccay, E.S.; Mateo, E.B., Statistics and Probability, First
Edition, Rex Book Store
Chua, Simon L., et al. 2016. “Hypothesis Testing Concerning Proportions.” In
21st Century Statistics and Probability, 41-45. Quezon City: Phoenix
Publishing House, Inc.
Malate, Jose S. 2018. “Hypothesis Tests for a Population Proportion. In
Statistics and Probability, 144-148. Sta. Ana, Manila: Vicarish
Publications and Trading, Inc.
Sevilla, Consuelo G. Research Methods. (C.M. Recto Avenue Manila,
Philippines: Rex Bookstore, 2009) 14 – 19.
Department of Education. Division of Pasig City
Department of Education – Region IV-A CALABARZON