(Chapter–14): Hypothesis Testing RISE Quantitative Methods

Chapter-14

HYPOTHESIS TESTING

INTRODUCTION
Hypothesis testing is a very important phase of statistical inference. It is a procedure which enables
us to decide on the basis of information obtained from sample data whether to accept or reject a
statement or an assumption about the value of a population parameter. Such a statement or
assumption which may or may not be true; called a statistical hypothesis. We accept the hypothesis
as being true, when it is supported by the sample data. We reject the hypothesis when the sample,
data fail to support it.
It is important to understand what we mean by the terms reject and accept in hypothesis testing.
The rejection of a hypothesis is to declare it false. The acceptance of a hypothesis is to conclude
that there is not sufficient evidence to reject it. Acceptance does not necessarily mean that the
hypothesis is true.
Null Hypothesis
It is a hypothesis which is to be tested for possible rejection under the assumption that it is true. It
is denoted by the H0.
Alternative Hypothesis
It is any other hypothesis which we accept when the null hypothesis is rejected it is denoted by H1.
Level of significance
It is the probability of rejecting the null hypothesis when it is true. OR
The probability of committing Type–I error is called level of significance.
Test Statistic
A sample statistic is a formula which provides a basis for testing a null hypothesis is called a test
statistic. The most commonly used test statistics are Z, T, 2 & F.
Critical Region / Rejection Region
A set of values which leads to reject null hypothesis is called critical region.
Non–critical Region / Acceptance Region
A set of values / (area) which leads to accept null hypothesis is called non–critical / acceptance
region.
Two–Tailed Test
If the critical region is located in both tails of the sampling distribution of test statistic is called two
tailed test.

(220)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

One–Tailed Test
If the critical region is located in one tail of the test statistic is called one–tailed test.

Type–T Error
Type–I error occurs if an investigator rejects a null hypothesis that is actually true.
Type–II Error
Type–II error occurs if an investigator accepts the null hypothesis that is actually false.
Level of Confidence
The probability of accepting a null hypothesis when it is true is called level of confidence.

Important MCQs
1) A procedure to determine whether or not as assumption about some parameter of a population
is supported by the information obtained from the observed random sample is known as:
A) Hypothesis testing B) Estimation
C) Inferential statistics D) Sampling distribution

2) A statement which is tested for possible rejection under the assumption that it is true is called:
A) Statistical hypothesis B) Null hypothesis
C) Alternative hypothesis D) None of these

3) The probability of rejecting the true null hypothesis is called:

A) Level of significance B) Test statistic
C) Level of confidence D) Type-II error

4) A significance level of 0.01 means:

A) If null hypothesis is rejected, there is a maximum chance of 1% that the decision may
be wrong
B) There is more than 99% chance that the null hypothesis is false
C) If the null hypothesis is accepted, there is at least1% chance that the decision may be
wrong
D) If type II error is made, there is 99% chance of making a type 1 error too

5) If the level of significance is increased, the area of the critical region will:
A) Increased B) Decreased
C) Remain unchanged D) None of these

6) The rule or formula that provides the basis for testing a null hypothesis is called:
A) Level of significance B) Critical region
C) Test statistic D) None of these

7) The area specified for the values which leads to reject null hypothesis is known as:
A) Test statistic B) Acceptance region
C) Critical region D) Level of significance

(221)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

8) Identify the critical value of z-test, if level of significance is 5% for a one tail test.
A) ± 1.96 B) ± 1.645
C) ± 2.33 D) ± 2.58

9) Rejection of the null hypothesis, when it is true is known as:

A) Level of confidence B) Type-I error
C) Type-II error D) Critical region

10) A failing student is passed by the examiner is an example of:

A) Level of confidence B) Type-I error
C) Type-II error D) Critical region

1. TESTING OF HYPOTHESIS ABOUT POPULATION MEAN

Testing Procedure
1. Formulate the null and alternative hypothesis
𝐻0 :  = 𝜇0 or 𝐻0 :  ≤ 𝜇0 or 𝐻0 :  ≥ 𝜇0
𝐻1 :  ≠ 𝜇0 or 𝐻1 :  > 𝜇0 or 𝐻1 :  < 𝜇0
2. Level of significance
 = 0.05, 0.01 or etc (Given)
3. The Test Statistic
x−
z= when  is known

n
x−
z= when  is unknown & n > 30
s
n
4. Critical Region
(a) If 𝐻1 :  ≠ 𝜇0 , then the C.R will be
z < − 𝑧𝛼 and z > 𝑧𝛼
2 2

(b) If 𝐻1 :  > 𝜇0 , then the C.R will be

z > z
(c) If 𝐻1 :  < 𝜇0 , then the C.R will be
z < – z
5. Computation
6. Conclusion
(a) If the calculated value of z falls in the C.R then we reject H0.
(b) If the calculated value of z does not fall in C.R then we accept H0.

(222)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

Example 01:
The average distance spent by each person commuting to the office each month in a large city is
asserted to be 460Km. A sample of 100 commuters showed a mean of 450Km and had a standard
deviation of 25Km. Does the sample support the assertion about the population at the 5%
significance level?
Solution:
1. 𝐻0 :  = 460
𝐻1 :  ≠ 460
2. Level of Significance
 = 0.05
3. The Test Statistic
x−
z=
s
n
4. Critical Region
z < –𝑧𝛼 and z >𝑧𝛼
2 2

z < – z0.025 and z > 0.025

z < – 1.96 and z > 1.96

5. Computation
450 − 460
z= =–4
25
100
6. Conclusion
Since the computed value of Z falls in the critical region. So, we reject H 0 at 5% level of
significance.

Example 02:
A cable manufacturer makes cable with a breaking strength of 700Kg with a standard deviation of
13Kg. The manufacturer has invented a different technique that it claims the results in stronger
cable. A random sample of 40 cables has a mean strength of 725Kg.
Solution:
1. 𝐻0 :  ≤ 700
𝐻1 :  > 700
2. Level of Significance
 = 0.05

(223)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

3. The Test Statistic

x−
z=

n

4. Critical Region
We reject 𝐻0
z > z
z > z0.05
z > 1.645

725 − 700
5. z= = 12.16
13
40
6. Since the computed value of Z falls in the critical region. So, we reject H0 at 5% level of
significance.

Important MCQs
11) Which of the following construction of null and alternative hypothesis is/are correct? Please
note that population mean in both cases is 36:
(i) The sample data provide evidence that the population mean is less than 36. Therefore,
Ho: 𝜇 < 36 H1: 𝜇 = 36
(ii) The sample data provide evidence that the population mean is greater than 36. Therefore,
Ho: 𝜇 = 36 H1: 𝜇 > 36
A) Only statement (i) is correct B) Only statement (ii) is correct
C) Both statements are correct D) Both statements are not correct

12) An LED light manufacturing company claims that the light produced by them have a mean life
time of 10.000 hours. Construct an appropriate null and alternative hypothesis:
A) 𝐻𝑜 : 𝜇 = 10,000 B) 𝑯𝒐 : 𝝁 ≥ 𝟏𝟎, 𝟎𝟎𝟎
𝐻1 : 𝜇 ≠ 10,000 𝑯𝟏 : 𝝁 < 𝟏𝟎, 𝟎𝟎𝟎
C) 𝐻𝑜 : 𝜇 ≤ 10,000 D) 𝐻𝑜 : 𝜇 = 10,000
𝐻1 : 𝜇 > 10,000 𝐻1 : 𝜇 > 10,000

13) If n > 30 and the variance of population mean is unknown, then the test statistic for the testing
the population mean must be:
A) t – test B) z – test
C) F – test D) 2 – test

(224)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

14) In which of the following case is the level of significance divided into two tails equally:
A) A manufacturer claims that his product last for more than 24 hours.
B) A milk bottling plant is expected to fill 250 ml of liquid in one bottle.
C) A manufacturer claims that defects in its production are less than 5%.
D) A school claims that their passing ratio is 75% or more

15) A pharmaceutical company claims that the amount of alcohol in each bottle of drug is 0.706mg.
A random sample of 38 bottles of that drug was tested and found to have mean alcohol content
of 0.705mg with a standard deviation of 0.02mg, if we test the company’s claim at 5% of level
of significance, which of the following statements will become true?
(1) Calculate value of Z is – 0.308
(2) Table value of Z is 1.96
(3) Claim is accepted
A) 1, 2 B) 2, 3
C) 1, 3 D) All are correct

16) Perform hypothesis testing on the following data and conclude whether to accept or reject null
hypothesis:
i) Claim: µ ≤ 8 ii) Population standard deviation = 0.2
iii) Sample size = 49 iv) Sample mean = 7.95
v) Significance level = 1%
A) Accept B) Reject
B) Cannot be determined as sample standard deviation is not given
C) Difficult to tell

2. TESTING OF HYPOTHESIS ABOUT POPULATION MEAN

Testing Procedure
1. Formulate the null and alternative hypothesis
𝐻0 :  = 𝜇0 or 𝐻0 :  ≤ 𝜇0 or 𝐻0 :  ≥ 𝜇0
𝐻1 :  ≠ 𝜇0 or 𝐻1 :  > 𝜇0 or 𝐻1 :  < 𝜇0
2. Level of significance
 = 0.05, 0.01 or etc (Given)
3. The Test Statistic
x−
t= when  is unknown & n  30
s
n
x ( x − x )2 1  2 (  x )2 
Where, x = , s= =
n n −1  x − 
n −1 n 

4. Critical Region
(a) If 𝐻1 :  ≠ 𝜇0 , then the C.R will be
t  −𝑡𝛼,(𝑛−1) , and t  𝑡𝛼,(𝑛−1)
2 2

(b) If 𝐻1 :  > 𝜇0 , then the C.R will be

t  𝑡𝛼,(𝑛−1)
(c) If 𝐻1 :  < 𝜇0 , then the C.R will be
t  – 𝑡𝛼,(𝑛−1)

(225)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

5. Computation
6. Conclusion
(a) If the computed value of t falls in the C.R, then we reject H0.
(b) If the compute value of t does not fall in the C.R then we accept H0.
Example 03:
An investigator suspects that the mean IQ level for a certain type of test is 110. He designs an
experiment where 5 volunteers each took the test and achieve the following scores. Do the scores
support his suspicion at the 95% confidence level?
Volunteer Test Score
1 135
2 103
3 129
4 96
5 121
Solution:
1. 𝐻0 :  = 110
𝐻1 :  ≠ 𝜇0
2. Level of Significance
 = 0.05
3. The Test Statistic
x−
t=
s
n
4. Critical Region
t  −𝑡𝛼,(𝑛−1) , and t  𝑡𝛼,(𝑛−1)
2 2
t  – t0.025, 4 and t  t0.025, 4
t  – 2.7764 and t  2.7764

5. Computation
x = 135, 103, 129, 96, 121
By calculator
Mode → Stat → 1 – Var → x → AC
135
103
129
96
121
→ Shift 1 → Var → x = 116.8
→ Shift 1 → Var → xn-1 (s) = 16.74

(226)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

So,
116.8 − 110
t= = 0.91
16.74
5
6. Since the computed value of t does not fall in the critical region. So, we accept H0 at 5% level
of significance.
Example 04:
A random sample of 25 hens from a normal population showed that the average laying is 272 eggs
per year with a variance of 625 eggs. The company claimed that the average laying is at least 285
eggs per year. Test the claim of the company at  = 0.05
Solution:
1. 𝐻0 :  ≥ 285
𝐻1 :  < 285
2. Level of Significance
 = 0.05
3. The Test Statistic
x−
t=
s
n
4. Critical Region
𝑡 ≤ 𝑡𝛼,(𝑛−1)
𝑡 ≤ 𝑡0.05,(25−1)
t  – 1.711

5. Computation
n = 25
s2 = 625  s = 25
x = 272
Now
272 − 285
t= = - 2.6
25
25
6. Conclusion:
Since the computed value of t falls in the critical region. So, we reject H0 at 5%.

(227)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

Important MCQs
17) t – test is only used in testing of hypothesis about population mean if:
A) 𝑛 ≤ 30 B)  is unknown
C) 𝜎 2 is known D) Both A and B

18) Perform hypothesis testing on the following data and conclude whether to accept or reject null
hypothesis:
i) Claim: µ ≤ 11 ii) Sample standard deviation = 0.2
iii) Sample size = 10 iv) Sample mean = 11.4
v) Significance level = 1%
A) Accept B) Reject
C) Cannot be determined as population standard deviation is not given
D) Difficult to tell

3. TESTING OF HYPOTHESIS ABOUT POPULATION PROPORTION

Testing Procedure
1. Formulate the null and alternative hypothesis
𝐻0 : 𝑝 = 𝑝0 or 𝐻0 : 𝑝 ≤ 𝑝0 or 𝐻0 : 𝑝 ≥ 𝑝0
𝐻1 : 𝑝 ≠ 𝑝0 or 𝐻1 : 𝑝 > 𝑝0 or 𝐻1 : 𝑝 < 𝑝0
2. Level of significance
 = 0.05, 0.01 or etc (Given)
3. The Test Statistic

Pˆ − P x
z= Where P̂ =
P(1 − P ) n
n
4. Critical Region
(a) If 𝐻1 : 𝑝 ≠ 𝑝0 , then the C.R will be
z < − 𝑧𝛼 and z > 𝑧𝛼
2 2

(b) If 𝐻1 : 𝑝 > 𝑝0 , then the C.R will be

z > z
(c) If 𝐻1 : 𝑝 < 𝑝0 , then the C.R will be
z < – z
5. Computation
6. Conclusion
(a) If the calculated value of z falls in the C.R then we reject H0.
(b) If the calculated value of z does not fall in C.R then we accept H0.

(228)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

Example 05:
An election candidate claims that 60% of the voters support him. A random sample of 2500 voters
showed that 1410 supported him. Test the candidates claim at 95% confidence level.
Given:
p = 60% = 0.60
n = 2500
x 1410
x = 1410 𝑃̂ = = = 0.564
n 2500
Solution:
1. 𝐻0 : 𝑝 = 𝑝0
𝐻1 : 𝑝 ≠ 𝑝0
2. Level of Significance:
 = 0.05
3. The Test Statistic:
𝑃̂−𝑃
z=
𝑃(1−𝑃)
√
𝑛

4. Critical Region:
z < –𝑍𝛼 and >𝑍𝛼
2 2

z < – z0.025 and z > z0.025

z < – 1.96 and z > 1.96

5. Computation:
0.564 − 0.60
z= = – 3.7
0.60(1 − 0.60)
2500
6. Conclusion:
Since the computed value of Z falls in the critical region. So, we reject H0 at 5%.

(229)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

Example 07:
An election candidate claims that at least 60% of the voters support him. A random sample of 2500
voters showed that 1410 supported him. Test the claims at 95% confidence level?

Solution:
1. 𝐻0 : 𝑝 ≥ 𝑝0
𝐻1 : 𝑝 < 𝑝0
2. Level of Significance:
 = 0.05
3. The Test Statistic:
𝑃̂ −𝑃
z= 𝑃(1−𝑃)
√
𝑛

4. Critical Region:
z < – z
z < – z0.05
z < – 1.645
5. Computation:
𝑃̂ −𝑃 x 1410
z= 𝑃̂ = = = 0.564
𝑃(1−𝑃) n 2500
√
𝑛

0.564 − 0.60
z= = - 3.7
0.60(1 − 0.60)
2500
6. Conclusion:
Since the computed value of z falls in the Critical Region. So, we reject H0 at 5% level of
significance.

Important MCQs
19) A basketball player has hit on 60% of his shots from the floor. If on the next 100 shots he makes
70 baskets, would you say that his shooting has improved using 5% level of significance?
A) 2.041 and 1.645 B) 4.753 and 1.96
C) 2.041 and 1.96 D) 4.753 and 1.645

20) Perform hypothesis testing on the following data and conclude whether to accept or reject null
hypothesis:
i) Claim: P < 0.7 ii) Sample proportion = 0.684
iii) Standard error of proportion = 0.0092 iv) Level of Significance = 10%
A) Accept B) Reject
C) Can’t be determined because population proportion is not given
D) Can’t be determined because sample size is not given.

(230)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

4. Testing of Hypothesis about the Difference between Two Population Means.

Testing Procedure:
1. Formulate the null and alternative hypothesis
H0 ∶ There is no significance difference between two population mean
H1 ∶ There is a significance difference between two population mean
OR
(i) H0 ∶ μ1 = μ2 and H1 ∶ μ1 ≠ μ2
(ii) H0 ∶ μ1 ≥ μ2 and H1 ∶ μ1 < μ2
(iii) H0 ∶ μ1 ≤ μ2 and H1 ∶ μ1 > μ2

OR
(i) H0 ∶ μ1 − μ2 = ∆0 and H1 ∶ μ1 − μ2 ≠ ∆0
(ii) H0 ∶ μ1 − μ2 ≥ ∆0 and H1 ∶ μ1 − μ2 < ∆0
(iii) H0 ∶ μ1 − μ2 ≤ ∆0 and H1 ∶ μ1 − μ2 > ∆0

2. Level of significance
 = 0.05, 0.01 or etc (Given)
3. The Test Statistic
(𝑋̅1 −𝑋̅2 )− (𝜇1 −𝜇2 ) (𝑋̅1 −𝑋̅2 )− ∆0
𝑍= OR 𝑍= when 1 and 2 are known
𝜎2 𝜎2 𝜎2 𝜎2
√ 1 + 2 √ 1 + 2
𝑛1 𝑛2 𝑛1 𝑛2

(𝑋̅1 −𝑋̅2 )− (𝜇1 −𝜇2 ) (𝑋̅1 −𝑋̅2 )− ∆0

𝑍= OR 𝑍= when 1 and 2 are unknown
𝑆 2
𝑆 2 𝑆2 𝑆 2
√ 1+ 2 √ 1+ 2 and n1 and n2 > 30
𝑛1 𝑛2 𝑛1 𝑛2

4. Critical Region
(a) If H1:   0, then the C.R will be
z < − 𝑧𝛼 and z > 𝑧𝛼
2 2

(b) If H1:  > 0, then the C.R will be

z > z
(c) If H1:  < 0, then the C.R will be
z < – z
5. Computation
6. Conclusion
(a) If the computed value of z falls in the C.R, then we reject H0.
(b) If the compute value of z does not fall in the C.R then we accept H0.

(231)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

Example 08:
Regular sugar tests for women and men in a city are provided the table below:
Gender Women Men
Sample 1900 1600
Sugar level on average 125 130
Standard deviation of sample 20 17.5

Can we statistically verify that sugar level for women is lower than men.
Solution:
1. H0 ∶ μ1 ≥ μ2 and H1 ∶ μ1 < μ2
2. Level of significance
 = 0.05
(𝑋̅1 −𝑋̅2 )− (𝜇1 −𝜇2 )
3. 𝑍 =
𝑆 𝑆 2 2
√ 1+ 2
𝑛1 𝑛2

4. Critical Region
We reject Ho if
z < − 𝑧𝛼
z < − 𝑧0.05
z < − 1.645 (True)
5. Computation
(125−130)−0
𝑧= 2 2
= − 7.93
√(20) +(17.5)
1900 1600

6. Conclusion
Since the computed value of z falls in the critical region So, we reject Ho at 5% level of
significance. And conclude that the sugar level of women is less than the men.

Example 09:
A random sample of size 36 from a normal population with variance 24 gave 𝑋̅1 = 15. A second
sample of size 28 from another normal population with variance 80 gave 𝑋̅2 = 13. Test H0 ∶ μ1 =
μ2 and H1 ∶ μ1 ≠ μ2 Let 𝛼 = 0.05.
Solution:
1. H0 ∶ μ1 = μ2 and H1 ∶ μ1 ≠ μ2
2. Level of significance
 = 0.05
(𝑋̅1 −𝑋̅2 )− (𝜇1 −𝜇2 )
3. 𝑍 =
𝜎 𝜎 2 2
√ 1+ 2
𝑛1 𝑛2

(232)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

4. Critical Region
We reject Ho if
z < −𝑧𝛼 and z >𝑧𝛼
2 2
z < −𝑧0.05 and z >𝑧0.05
2 2
z < −𝑧0.025 and z >𝑧0.025
z < −1.96 and z > 1.96
5. Computation
(15−13)−0
𝑧= 24 80
= 1.06
√ +
36 28

6. Conclusion
Since the computed value of z does not fall in the critical region So, we accept Ho at 5%
level of significance.

5. Testing of Hypothesis about the Difference between Two Population Means. (when
𝝈𝟏 = 𝝈𝟐 but unknown, and 𝒏𝟏 and 𝒏𝟐 ≤ 30)
Testing Procedure:
1. Formulate the null and alternative hypothesis
H0 ∶ There is no significance difference between two population mean
H1 ∶ There is a significance difference between two population mean
OR
(i) H0 ∶ μ1 = μ2 and H1 ∶ μ1 ≠ μ2
(ii) H0 ∶ μ1 ≥ μ2 and H1 ∶ μ1 < μ2
(iii) H0 ∶ μ1 ≤ μ2 and H1 ∶ μ1 > μ2
OR
(i) H0 ∶ μ1 − μ2 = ∆0 and H1 ∶ μ1 − μ2 ≠ ∆0
(ii) H0 ∶ μ1 − μ2 ≥ ∆0 and H1 ∶ μ1 − μ2 < ∆0
(iii) H0 ∶ μ1 − μ2 ≤ ∆0 and H1 ∶ μ1 − μ2 > ∆0
2. Level of significance
 = 0.05, 0.01 or etc (Given)
3. The Test Statistic
(𝑋̅1 −𝑋̅2 )− (𝜇1 −𝜇2 ) (𝑋̅1 −𝑋̅2 )− ∆0
𝑡= 1 1
OR 𝑡= 1 1
when 1 and 2 are unknown
𝑠𝑝 √ + 𝑠𝑝 √ +
𝑛1 𝑛2 𝑛1 𝑛2

but equal and 𝑛1 and 𝑛2 ≤ 30

Where,
∑(𝑋1 −𝑋̅1 )2 + ∑(𝑋1 −𝑋̅2 )2
i) 𝑠𝑝 = √ 𝑛1 +𝑛2 −2

(𝑛1 −1)𝑠12 +(𝑛2 −1)𝑠12

ii) 𝑠𝑝 = √ 𝑛1 +𝑛2 −2
∑𝑋 ∑𝑋
𝑋̅1 = 𝑛 1 , 𝑋̅2 = 𝑛 2
1 2

∑(𝑋1 −𝑋̅1 )2 ∑(𝑋2 −𝑋̅2 )2

𝑠12 = 𝑛1 −1
𝑠12 = 𝑛2 −1

(233)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

4. Critical Region
(a) If H1 ∶ μ1 ≠ μ2 , then the C.R will be
t  −𝑡𝛼,(𝑛 , and t  𝑡𝛼,(𝑛
2 1 +𝑛2 −2) 2 1 +𝑛2 −2)

(b) If H1 ∶ μ1 > μ2 , then the C.R will be

t  𝑡𝛼,(𝑛1 +𝑛2 −2)
(c) If H1 ∶ μ1 < μ2 , then the C.R will be
t  – 𝑡𝛼,(𝑛1 +𝑛2 −2)
5. Computation
6. Conclusion
(a) If the computed value of t falls in the C.R, then we reject H0.
(b) If the compute value of t does not fall in the C.R then we accept H0.

Example 10:
A feeding test is conducted on a herd of 25 milking cows to compare two diets, one of dewatered
alfalfa and the other of field-wilted alfalfa. Dewatered alfalfa has an economic advantage in that its
mechanical processing produces a liquid protein-rich by product that can be used to supplement
the feed of other animals. A sample of 12 cows randomly selected from the herd are fed dewatered
alfalfa; the remaining 13 cows are fed field-wilted alfalfa. From observations made over a three-
week period, the average daily milk production in pounds recorded for each cow is:
Dewatered alfalfa 35 47 55 29 40 39 32 41 42 57 51 39
Field-wilted alfalfa 44 44 56 46 47 38 58 53 49 35 46 30 41
Do the data strongly indicate that the milk yield is less with dewatered alfalfa than the field-wilted
alfalfa? Test at 𝛼 = 0.05.

Solution:
1. Formulate the null and alternative hypothesis
H0 ∶ μ1 ≥ μ2 and H1 ∶ μ1 < μ2

2. Level of significance
 = 0.05
3. The Test Statistic
(𝑋̅1 −𝑋̅2 )− (𝜇1 −𝜇2 )
𝑡= 1 1
𝑠𝑝 √ +
𝑛1 𝑛2

4. Critical Region
We reject Ho if
t  – 𝑡𝛼,(𝑛1 +𝑛2 −2)
t  – 𝑡0.05,(12+13−2)
t  – 1.714

(234)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

5. Computation
(𝑛1 −1)𝑠12 +(𝑛2 −1)𝑠12
𝑠𝑝 = √
𝑛1 +𝑛2 −2
By Calculator
𝑋̅1 = 42.25, 𝑋̅2 = 45.15
𝑠12 = 76.38 𝑠22 = 64.14
(12−1)76.38+(13−1)64.14
𝑠𝑝 = √ 12+13−2
= 8.366

Now,
(42.25−45.15)− 0
𝑡= 1 1
= − 0.87
(8.366) √ +
12 13

6. Conclusion
Since the computed value of t does not fall in the critical region So, we accept Ho at 5% level
of significance.
6. Testing of Hypothesis about the Difference between Two Population Means.
(Dependent Samples, Paired Observations (n1 = n2 = n), n ≤ 30)
Testing Procedure:
1. Formulate the null and alternative hypothesis ∵ 𝜇𝐷 = 𝜇1 − 𝜇2
(i) H0 ∶ μD = 0 and H1 ∶ μD ≠ 0
(ii) H0 ∶ μD ≥ 0 and H1 ∶ μD < 0
(iii) H0 ∶ μD ≤ 0 and H1 ∶ μD > 0

2. Level of significance
 = 0.05, 0.01 or etc (Given)
3. The Test Statistic
𝑑̄ −𝜇𝐷
t= sd
√𝑛
∑𝑑 ∑(𝑑−𝑑̅ )2 1 (∑ 𝑑)2
where 𝑑 = 𝑥 − 𝑦, 𝑑̅ = , 𝑠𝑑 = = (∑ 𝑑2 − )
𝑛 𝑛−1 𝑛−1 𝑛

4. Critical Region
(a) If 𝐻1 :  ≠ 𝜇0 , then the C.R will be
t  −𝑡𝛼,(𝑛−1) , and t  𝑡𝛼,(𝑛−1)
2 2

(b) If 𝐻1 :  > 𝜇0 , then the C.R will be

t  𝑡𝛼,(𝑛−1)
(c) If 𝐻1 :  < 𝜇0 , then the C.R will be
t  – 𝑡𝛼,(𝑛−1)
5. Computation
6. Conclusion
(a) If the computed value of t falls in the C.R, then we reject H0.
(b) If the compute value of t does not fall in the C.R then we accept H0.

(235)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

Example 11:
An investigator suspects that there is a difference between two ways of assessing IQ. He designs
an experiment where 5 volunteers each take both tests and achieve the following scores. Do the
scores support his suspicion at the 95% confidence level?
Volunteer Test-A Marks Test-B Marks
1 135 125
2 103 102
3 129 117
4 96 94
5 121 121

Solution:
1. Formulate the null and alternative hypothesis
H0 ∶ μD = 0 and H1 ∶ μD ≠ 0
2. Level of significance
 = 0.05
3. The Test Statistic
𝑑̄ −𝜇𝐷
t= sd
√𝑛
4. Critical Region
We reject Ho if:
t  −𝑡𝛼,(𝑛−1) and t  𝑡𝛼,(𝑛−1)
2 2

t  −𝑡0.05,(5−1) and t  𝑡0.05,(5−1)

2 2

t  −𝑡0.025,(4) and t  𝑡0.025,(4)

t  −2.7764 and t  2.7764
5. Computation:
By calculator
𝑑̅ = 5, 𝑠𝑑 = 5.57
5−0
t= 5.57 =2
√5
6. Conclusion:
Since the computed value of t does not fall in the critical region. So, we accept Ho at 5% level
of significance.

(236)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

7. Testing of Hypothesis about the Difference between Two Population Proportion.

Testing Procedure:
1. Formulate the null and alternative hypothesis

(i) H0 ∶ p1 = p2 and H1 ∶ p1 ≠ p2 (H0 ∶ p1 − p2 = 0 and H1 ∶ p1 − p2 ≠ 0)

(ii) H0 ∶ p1 ≥ p2 and H1 ∶ p1 < p2
(iii) H0 ∶ p1 ≤ p2 and H1 ∶ p1 > p2

OR
(i) H0 ∶ p1 − p2 = ∆0 and H1 ∶ p1 − p2 ≠ ∆0 (∆0 = 𝑎𝑛𝑦 𝑣𝑎𝑙𝑢𝑒 (0.01 − 0.99))
(ii) H0 ∶ p1 − p2 ≥ ∆0 and H1 ∶ p1 − p2 < ∆0
(iii) H0 ∶ p1 − p2 ≤ ∆0 and H1 ∶ p1 − p2 > ∆0

2. Level of significance
 = 0.05, 0.01 or etc (Given)

3. The Test Statistic

(𝑝̂1 −𝑝̂2 )−∆0 𝑋1 𝑋2
𝑍= ̂ 𝑞
𝑝 ̂ ̂ 𝑞
𝑝 ̂
Where, 𝑝̂1 =
𝑛1
𝑎𝑛𝑑 𝑝̂ 2 =
𝑛2
√ 1 1 + 2 2
𝑛1 𝑛2

(𝑝̂1 −𝑝̂2 )− (p1 − p2 ) 𝑛1 𝑝̂1 +𝑛2 𝑝̂2

𝑍= 1 1
Where, 𝑝̂𝑐 =
𝑛1 +𝑛2
and 𝑞̂𝑐 = 1 − 𝑝̂𝑐
√𝑝̂𝑐 𝑞̂𝑐 (𝑛 + )
𝑛2
1

4. Critical Region
(a) If H1 ∶ p1 ≠ p2 , then the C.R will be
z < 𝑧𝛼 and z >𝑧𝛼
2 2

(b) If H1 ∶ p1 > p2 , then the C.R will be

z > z
(c) If H1 ∶ p1 < p2 , then the C.R will be
z < – z
5. Computation
6. Conclusion
(a) If the computed value of z falls in the C.R, then we reject H0.
(b) If the compute value of z does not fall in the C.R then we accept H0.

Example 12:
A health official claims that citizens of Peshawar are fitter than citizens of Gujrat. 96 out of 200
(45%) citizens in Peshawar (selected at random) passed a standard fitness test. 84 out of 200
(42%) citizens in Gujrat (selected at random) passed the same test. Test the official’s claim at the
95% confidence level.
1. H0 ∶ p1 ≤ p2 and H1 ∶ p1 > p2
2. Level of significance
 = 0.05

(237)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

3. The Test Statistics

(𝑝̂1 −𝑝̂2 )− ∆0
𝑍= 1 1
√𝑝̂𝑐 𝑞̂𝑐 (𝑛 + 𝑛 )
1 2

4. Critical Region
We reject Ho if
z > z
z > z0.05
z > 1.645
5. Computation
(0.48−0.42)− 0
𝑍= 1 1
= 1.20
√(0.45 × 0.55)( + )
200 200

6. Conclusion
Since the computed value of z does not fall in the critical region So, we accept Ho at 5% level
of significance.

Example 13:
A candidate for mayor in a large city believes that he appeals to at least 10 percent more of the
women voters than the men voters. He hires the services of a poll-taking organization, and they
find that 62 of 100 women interviewed support the candidate, and 69 of 150 men support him. At
the 0.05 significance level, is the hypothesis accepted or rejected?

1. H0 ∶ p1 − p2 ≤ 0.10 and H1 ∶ p1 − p2 > 0.10

2. Level of significance
 = 0.05

3. The Test Statistic

(𝑝̂1 −𝑝̂2 )−∆0
𝑍= ̂ 𝑞
𝑝 ̂ ̂ 𝑞
𝑝 ̂
√ 1 1+ 2 2
𝑛1 𝑛2

4. Critical Region
We reject Ho if
z > z
z > z0.05
z > 1.645

5. Computation
(𝑝̂1 −𝑝̂2 )−∆0
𝑍= ̂ 𝑞
𝑝 ̂ ̂ 𝑞
𝑝 ̂
√ 1 1+ 2 2
𝑛1 𝑛2

𝑋 𝑋
𝑝̂1 = 𝑛1 𝑎𝑛𝑑 𝑝̂2 = 𝑛2
1 2

(238)
(Chapter–14): Hypothesis Testing RISE Quantitative Methods

62 69
𝑝̂1 = 100 = 0.62 𝑎𝑛𝑑 𝑝̂2 = 150 = 0.46

𝑞̂1 = 0.38 𝑎𝑛𝑑 𝑞̂2 = 0.54

Now
(0.62−0.46)−0.10
𝑍= = 0.95
(0.62 ×0.38) (0.46 ×0.54)
√ +
100 150

6. Conclusion
Since the computed value of z does not fall in the critical region So, we accept Ho at 5% level
of significance.

(239)