Hypothesis Testing Involving

One Population
Learning Objectives:
• identifies the appropriate form of the test-statistic when the population
variance is assumed to be known and the population variance is assumed
to be unknown
• identifies the appropriate rejection region for a given level of significance
when the population variance is assumed to be known and the population
variance is assumed to be unknown
• computes for the test-statistic value (population mean and population
proportion)
• draws conclusion about the population mean and population proportion
based on the test-statistic value and the rejection region
• solves problems involving test of hypothesis on the population mean and
population proportion.
Steps in Hypothesis
Testing
1. Formulate the null and alternative
hypotheses.
2. Specify the level of significance and
choose the appropriate statistical tool.
3. Compute the test statistic.
4. Establish the critical region.
5. Make a decision and a conclusion
Alternative Hypotheses:

𝑯𝒂 : 𝝁 ≠ 𝝁𝟎 Two−tailed test

𝑯𝒂 : 𝝁 > 𝝁𝟎 Right−tailed test

𝑯𝒂 : 𝝁 < 𝝁𝟎 Left−tailed test

z-Test for One Population Mean and σ is
Known
Establishing Critical Region
Example 1. For each of the following, identify the
statistical test and indicate whether the decision
is to reject or do not reject the null hypothesis.

(a) 𝒛 < −𝟏. 𝟗𝟔, 𝒛 = −𝟏. 𝟎𝟖 𝑳𝒆𝒇𝒕 − 𝑻𝒂𝒊𝒍𝒆𝒅 𝑻𝒆𝒔𝒕

𝑯𝟎 : 𝒛 ≥ −𝟏. 𝟗𝟔 −𝟏. 𝟎𝟖 ≥ −𝟏. 𝟗𝟔 𝑻𝒓𝒖𝒆

𝑫𝒐 𝒏𝒐𝒕 𝒓𝒆𝒋𝒆𝒄𝒕 𝑵𝒖𝒍𝒍 𝑯𝒚𝒑𝒐𝒕𝒉𝒆𝒔𝒊𝒔

 Example 1. For each of the following, identify the
statistical test and indicate whether the decision
is to reject or do not reject the null hypothesis.

(b) 𝐳 < −𝟏. 𝟔𝟒𝟓 𝒐𝒓 𝒛 > 𝟏. 𝟔𝟒𝟓, 𝒛 = 𝟏. 𝟗𝟏 𝑻𝒘𝒐 − 𝑻𝒂𝒊𝒍𝒆𝒅 𝑻𝒆𝒔𝒕

𝑯𝟎 : 𝒛 ≤ 𝟏. 𝟔𝟒𝟓 1. 𝟗𝟏 ≤ 𝟏. 𝟔𝟒𝟓 𝑭𝒂𝒍𝒔𝒆

𝑹𝒆𝒋𝒆𝒄𝒕 𝑵𝒖𝒍𝒍 𝑯𝒚𝒑𝒐𝒕𝒉𝒆𝒔𝒊𝒔

Example 1. For each of the following, identify the
statistical test and indicate whether the decision
is to reject or do not reject the null hypothesis.

(c) 𝒛 > 𝟐. 𝟓𝟕𝟔, 𝒛 = 𝟏. 𝟓𝟒 𝑹𝒊𝒈𝒉𝒕 − 𝑻𝒂𝒊𝒍𝒆𝒅 𝑻𝒆𝒔𝒕

𝑯𝟎 : 𝒛 ≤ 𝟐. 𝟓𝟕𝟔 𝟏. 𝟓𝟒 ≤ 𝟐. 𝟓𝟕𝟔 𝑻𝒓𝒖𝒆

𝑫𝒐 𝒏𝒐𝒕 𝒓𝒆𝒋𝒆𝒄𝒕 𝑵𝒖𝒍𝒍 𝑯𝒚𝒑𝒐𝒕𝒉𝒆𝒔𝒊𝒔

 Example 2. Automotive
engineers tested the gas mileage STEP 1. Formulate the null
(in kilometers per liter or km/L) and alternative hypotheses.
of a passenger car model from a
certain car company. A random 𝐻0 : The average gas mileage of the
sample of 35 cars resulted in to passenger car model is 16km/L
mean gas mileage of 15km/L. 𝜇 = 16𝑘 𝑚Τ𝐿
The standard deviation of all
cars, obtained from the previous 𝐻𝐴 : The average gas mileage of the
test, is 2.5km/L. The car passenger car model is not 16km/L
company claims that the average
gas mileage of the passenger car 𝜇 ≠ 16𝑘 𝑚Τ𝐿
model is 16km/L. Test if the
claim is valid at a 5% level of
significance.
 Example 2. Automotive
STEP 2. Specify the level of significance
engineers tested the gas mileage and choose the appropriate statistical tool.
(in kilometers per liter or km/L)
𝐻0 : The average gas mileage of the
of a passenger car model from a
passenger car model is 16km/L
certain car company. A random 𝜇 = 16𝑘 𝑚Τ𝐿
sample of 35 cars resulted in to 𝐻𝐴 : The average gas mileage of the
mean gas mileage of 15km/L. passenger car model is not 16km/L
The standard deviation of all 𝜇 ≠ 16𝑘 𝑚Τ𝐿
cars, obtained from the previous 𝟓% 𝒍𝒆𝒗𝒆𝒍 𝒐𝒇 𝒔𝒊𝒈𝒏𝒊𝒇𝒊𝒄𝒂𝒏𝒄𝒆
test, is 2.5km/L. The car
company claims that the 𝒐𝒓 𝜶 = 𝟎. 𝟎𝟓
average gas mileage of the 𝑻𝒘𝒐 − 𝒕𝒂𝒊𝒍𝒆𝒅 𝒕𝒆𝒔𝒕
passenger car model is 16km/L.
Test if the claim is valid at a 5%
level of significance.
 STEP 3. Compute the test statistic.
Illustrative Example 2. 𝐻0 : The average gas mileage of the
Automotive engineers tested passenger car model is 16km/L
the gas mileage (in kilometers 𝜇 = 16𝑘 𝑚Τ𝐿
per liter or km/L) of a passenger 𝐻𝐴 : The average gas mileage of the
car model from a certain car passenger car model is not 16km/L
company. A random sample of 35 𝜇 ≠ 16𝑘 𝑚Τ𝐿

cars resulted in to mean gas 𝜶 = 𝟎. 𝟎𝟓, 𝑻𝒘𝒐 − 𝒕𝒂𝒊𝒍𝒆𝒅 𝒕𝒆𝒔𝒕

mileage of 15km/L. The standard 𝒙lj = 𝟏𝟓𝒌 𝒎Τ𝑳 , 𝝈 = 𝟐. 𝟓𝒌 𝒎Τ𝑳 ,

deviation of all cars, obtained 𝒏 = 𝟑𝟓 𝒄𝒂𝒓𝒔, 𝝁𝟎 = 𝟏𝟔𝒌𝒎/𝑳
from the previous test, is
2.5km/L. The car company 𝑥lj − 𝜇0 15 − 16 −1
𝑧= 𝜎 = =
claims that the average gas 2.5 2.5
mileage of the passenger car 𝑛 35 5.9160 …
model is 16km/L. Test if the
claim is valid at a 5% level of −1
𝑧= 𝒛 = −𝟐. 𝟑𝟕
significance. 0.4225 …
 Example 2. Automotive
STEP 4. Establish the critical region.
engineers tested the gas mileage
𝐻0 : The average gas mileage of the
(in kilometers per liter or km/L) passenger car model is 16km/L
of a passenger car model from a 𝜇 = 16𝑘 𝑚Τ𝐿
certain car company. A random 𝐻𝐴 : The average gas mileage of the
sample of 35 cars resulted in to passenger car model is not 16km/L
mean gas mileage of 15km/L. 𝜇 ≠ 16𝑘 𝑚Τ𝐿
The standard deviation of all 𝜶 = 𝟎. 𝟎𝟓, 𝑻𝒘𝒐 − 𝒕𝒂𝒊𝒍𝒆𝒅 𝒕𝒆𝒔𝒕 𝒛 = −𝟐. 𝟑𝟕
cars, obtained from the previous 𝐳 < −𝟏. 𝟗𝟔 𝒐𝒓 𝒛 > 𝟏. 𝟗𝟔
test, is 2.5km/L. The car
company claims that the
average gas mileage of the
passenger car model is 16km/L.
Test if the claim is valid at a 5%
level of significance.
 Example 2. Automotive STEP 5. Make a decision and a conclusion
engineers tested the gas mileage 𝐻0: The average gas mileage of the passenger car model is 16km/L
𝜇 = 16𝑘 𝑚Τ𝐿
(in kilometers per liter or km/L)
of a passenger car model from a 𝐻𝐴 : The average gas mileage of𝜇the passenger car model is not 16km/L
≠ 16𝑘 𝑚Τ𝐿
certain car company. A random 𝜶 = 𝟎. 𝟎𝟓, 𝑻𝒘𝒐 − 𝒕𝒂𝒊𝒍𝒆𝒅 𝒕𝒆𝒔𝒕 𝐳 < −𝟏. 𝟗𝟔 𝒐𝒓 𝒛 > 𝟏. 𝟗𝟔
sample of 35 cars resulted in to 𝒛 = −𝟐. 𝟑𝟕
mean gas mileage of 15km/L.
The standard deviation of all
cars, obtained from the previous
test, is 2.5km/[Link] car
company claims that the average
gas mileage of the passenger car
model is 16km/[Link] if the Since the z-test statistic -2.37 falls in the critical
claim is valid at a 5% level of region, reject the null hypothesis. Therefore, there is
significance. sufficient evidence at the 5% level of significance
that the average gas mileage of the passenger car
model is significantly different from 16km/L.
t-Test for One Population Mean: σ is unknown
 Illustrative Example 3. Angel 𝑯𝟎 : The average grade in the mathematics
heard that the average in the
of the class is at least 88%, that is, 𝜇 ≥ 88%
mathematics of her class is at
least 88%. She was not 𝑯𝒂 : The average grade in the mathematics of
convinced by this, and so the class is less than 88%, that is, 𝜇 < 88%
decided to use hypothesis 𝛼 = 0.05, directional, left-tailed
testing to check if this claim test
was true. She got a random
sample of 10 classmates who 𝑥ҧ = 859 = 85.9
gave their grades in 10
mathematics as follows: 90, 93,
85, 77, 88, 80, 78, 83, 95, 90.
Assume that the distribution of
the grades is normal. Based on
this sample data, what would
Angel’s conclusion be on the
average grade in the
mathematics of her class?
 Example 3. Angel heard that 𝑯𝟎 : The average grade in the mathematics of the class
the average in the mathematics is at least 88%, that is, 𝜇 ≥ 88%
of her class is at least 88%. She 𝑯𝒂 : The average grade in the mathematics of the class
was not convinced by this, and is less than 88%, that is, 𝜇 < 88%
so decided to use hypothesis 𝛼 = 0.05, directional, left-tailed test 𝑥ҧ = 85.9
testing to check if this claim was𝒙 𝒙−𝒙𝝁 𝒙𝒙−−𝝁 𝝁 𝟐 𝒙 −𝒙 𝟖𝟖 𝟐
−𝝁 𝟐
true. She got a random sample 90 902 2
4 4
of 10 classmates who gave their 93 935 25
5 σ25𝒙−𝝁 𝟐
𝒔=
grades in mathematics as
85 85
-3 -3
9 9𝒏
follows: 90, 93, 85, 77, 88, 80, 78,
83, 95, 90. Assume that the 77 -11
77 121
-11 121
𝟒𝟎𝟏
distribution of the grades is 88 880 0 𝒔= 0
𝟏𝟎
normal. Based on this sample80 80
-8 64
-8 64
data, what would Angel’s 78 -10
78 100
-10 100
𝒔 = 𝟒𝟎. 𝟏
conclusion be on the average83 83
-5 25
-5 25
grade in the mathematics of her 95 957 49
7 𝒔 = 49
𝟔. 𝟑𝟑
class? 90 902 2
4 4
401 401
 Example 3. Angel heard that 𝑯𝟎 : The average grade in the mathematics of the class
the average in the mathematics is at least 88%, that is, 𝜇 ≥ 88%
of her class is at least 88%. She 𝑯𝒂 : The average grade in the mathematics of the class
was not convinced by this, and is less than 88%, that is, 𝜇 < 88%
so decided to use hypothesis 𝛼 = 0.05, directional, left-tailed test
testing to check if this claim was 𝑥ҧ = 85.9 𝑠 = 6.33 𝜇0 = 88 𝑛 = 10 𝑠𝑡𝑢𝑑𝑒𝑛𝑡𝑠
true. She got a random sample
of 10 classmates who gave their 𝑥ҧ − 𝜇0 85.9 − 88 −2.1
grades in mathematics as 𝑡= 𝑠 = =
6.33 6.33
follows: 90, 93, 85, 77, 88, 80, 78,
83, 95, 90. Assume that the
𝑛 10 3.1622 …
distribution of the grades is
normal. Based on this sample −2.1
data, what would Angel’s 𝑡= 𝒕 = −𝟏. 𝟎𝟓
conclusion be on the average 2.0017 …
grade in the mathematics of her
class?
 Example 3. Angel heard 𝑯𝟎 : The average grade in the mathematics of the class is at least
that the average in the 88%, that is, 𝜇 ≥ 88%
mathematics of her class is 𝑯𝒂 : The average grade in the mathematics of the class is less than
at least 88%. She was not 88%, that is, 𝜇 < 88%
convinced by this, and so 𝛼 = 0.05, directional, left-tailed test 𝑥ҧ = 85.9 𝜇0 = 88 𝑠 = 6.3
decided to use hypothesis 𝑛 = 10 𝑠𝑡𝑢𝑑𝑒𝑛𝑡𝑠
testing to check if this claim 𝑑𝑓 = 9, 𝛼 = 0.05 𝒕 < −𝟏. 𝟖𝟑𝟑
𝒕 = −𝟏. 𝟎𝟓
was true. She got a random
sample of 10 classmates
who gave their grades in
mathematics as follows: 90,
93, 85, 77, 88, 80, 78, 83, 95,
90. Assume that the
distribution of the grades is Since the computed t-test statistic value -1.05 does not fall
normal. Based on this under the critical region, do not reject the null hypothesis.
sample data, what would Therefore, there is no sufficient evidence from the sample
Angel’s conclusion be on the
average grade in the
data that the average grade in Mathematics of the class is
mathematics of her class? significantly lower than 88 %.
Hypothesis testing for One Population
Proportion p
 Example 4. A survey 𝐻𝑎 : The proportion of people who regard global
was conducted to
determine the opinions warming as a serious problem is significantly higher
of people on global than 60%. That is, 𝑝 > 60%
warming. In a random 𝐻0 : The proportion of people who regard global
sample of 150 people, warming as a serious problem is at most 60%. That
108 think that global is, 𝑝 ≤ 60% 𝛼 = 0.01, 𝑅𝑖𝑔ℎ𝑡 − 𝑡𝑎𝑖𝑙𝑒𝑑 𝑡𝑒𝑠𝑡
warming is a serious 108
world problem. Is there 𝑝= = 0.72 0.72 − 0.6 0.12
sufficient evidence that 150 𝑧= =
𝑝0 = 60% = 0.6 0.6 0.4 0.24
the proportion of
people who regard 𝑞0 = 1 − 0.6 = 0.4 150 150
global warming as a 𝑛 = 150 0.12 0.12
serious problem is 𝑧= =
𝑝 − 𝑝0 0.0016 0.04
significantly higher than 𝑧=
60%? use 1% level of 𝑝0 𝑞0 𝒛=𝟑
significance. 𝑛
Example 4. A survey 𝐻𝑎 : The proportion of people who regard global warming as a
was conducted to serious problem is significantly higher than 60%. That is, 𝑝 >
determine the 60%
opinions of people 𝐻0 : The proportion of people who regard global warming as a
on global warming. serious problem is at most 60%. That is, 𝑝 ≤ 60%
In a random sample 𝛼 = 0.01, 𝑅𝑖𝑔ℎ𝑡 − 𝑡𝑎𝑖𝑙𝑒𝑑 𝑡𝑒𝑠𝑡 𝑝 = 0.72 𝑝0 = 0.6 𝑞0 = 0.4 𝑛 = 150 𝒛=𝟑
of 150 people, 108 𝒛 > 𝟐. 𝟑𝟐𝟔
think that global
warming is a serious
world problem. Is
there sufficient
evidence that the
proportion of people
who regard global
warming as a serious Since the z-test statistic 3 falls in the critical region, reject the null
problem is hypothesis. Therefore, there is sufficient evidence at the 1% level of
significantly higher
than 60%? use 1% significance that the proportion of people who regard global warming
level of significance. as a serious problem is significantly higher than 60%.