CHAPTER 6

TESTING OF HYPOTHESES

Introduction
Testing of hypothesis is mainly concerned with addressing questions arising from experimental
and observational research such as:
• Is the mean age of women in business 35 years?
• Is there a significant difference in the viral loads between males and females infected
with HIV/AIDS?
• Is there any significant difference in the production levels of 4 maize hybrids in
Zimbabwe?

The objective of hypothesis testing is to decide or draw conclusions, based on information

gathered from a sample, on the true values of population parameters.

Definition: A statistical hypothesis is a claim, belief or suspicion about the parameter of a

population or a statistical model for a data generating process or system.

Definition: The null hypothesis a statistical hypothesis denoted by 𝑯𝟎 is a statistical

hypothesis which states that there is no change in the level of the population variable we are
interested in.

The null hypothesis is a claim that is established for the purpose of testing. This claim can
either be rejected or not rejected. If there is sufficient evidence to reject the null hypothesis
then the alternative hypothesis is accepted.

Definition: The alternative hypothesis or research hypothesis is a statistical hypothesis

denoted by 𝑯𝟏 which says that there is change in the level of the population variable we are
interested in.

A statistical test of hypothesis is a procedure that is used in conjunction with sample data to
decide whether to reject or not to reject the null hypothesis.
Definition: A test statistic, T, is a calculation made from sample data whose value is used
as a basis for deciding whether or not 𝑯𝟎 should be rejected.

Definition: A critical region or rejection region of a test is a set C chosen so that if the
value of the test statistic falls in it, then 𝑯𝟎 is rejected.

A critical value is a value that separates the acceptance region from the rejection region, and
these values are usually obtained from some statistical tables.

Type I and Type II Errors

Definition: In a statistical test, if 𝑯𝟎 is rejected when it is it is in fact true, the error so made
is called a Type I error. The probability of making this error is called the Significance level
and is usually denoted by 𝜶, i.e.
𝛼 = 𝑃(𝑇 ∊ 𝐶| 𝐻0 ) = 𝑃(𝑟𝑒𝑗𝑒𝑐𝑡 𝐻0 |𝐻0 𝑖𝑠 𝑡𝑟𝑢𝑒).

The significance level is usually set at 5%, for one to have a $95% confidence on the results
one obtains.

Definition: In a statistical test, if 𝑯𝟎 is not rejected or retained when it is it is in fact false, the
error so made is called a Type II error. The probability of making this error is usually denoted
by 𝛽 and it is given by
𝛽 = 𝑃(𝑇 ∊ 𝐴| 𝐻1 ) = 𝑃(𝑎𝑐𝑐𝑒𝑝𝑡 𝐻0 |𝐻0 𝑖𝑠 𝑓𝑎𝑙𝑠𝑒)

where 𝐴 is the compliment of set C, i.e. 𝐴 = 𝐶′,

the ecceptance region is a compliment of and 1 − 𝛽 is called the power of a test. The graph
of 1 − 𝛽 against 𝛽 is called the Operating characteristic curve.

Example

Let 𝑋1 be a random sample of size 1 from a population with a probability distribution

𝑓(𝑥) = 𝑥 𝜃−1 𝑒 −𝑥 , 𝑥 > 0, 𝜃 ∊ {1, 2}.
and suppose that one wants to test the hypothesis
 𝐻0 : 𝜃 = 1 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜃 = 2

using 𝑋1 as the test statistic and 𝐶 = [1.6, ∞) = 1.6 ≤ 𝑥 ≤ ∞ as the rejection region.
Calculate
(a) the probability of making a type I error, the significance level of the test,
(b) the probability of type II error and
(c) power of the test.

Solution

(a) ∝= 𝑃(𝑋 > 1.6|𝜃 = 1)

∞
= ∫ 𝑒 −𝑥 𝑑𝑥
1.6

= 𝑒 −1.6
= 0.2019.

(b) 𝛽 = 𝑃(𝐴|𝐻1 )
= 𝑃(𝑋 ≤ 1.6|𝜃 = 2)
1.6
= ∫ 𝑥𝑒 𝑥 𝑑𝑥
0

= 1 − 𝑒 −1.6 − 1.6𝑒 −1.6

= 0.4751
(c) 𝑃𝑜𝑤𝑒𝑟 𝑜𝑓 𝑎 𝑡𝑒𝑠𝑡 = 1 − 𝛽 = 0.5249

In most computer based tests of hypothesis, the critical values obtained from tables would not
be readily available, but we make the decision of either rejecting or not rejecting 𝐻0 based on
what we call a p-value.

Definition: P-value is the lowest level of significance at which the null hypothesis can be
rejected. Consequently, we reject 𝐻0 at the ∝ level, if 𝑝 − 𝑣𝑎𝑙𝑢𝑒 <∝.
Exercise

1. Suppose you are to verify the claim that 𝜇 > 20 on the basis of a random sample of size
58 and you are given that 𝜎 = 5.2.
(a) If you set your rejection region to be 𝑅: 𝑋̅ ≥ 21.15, what is the probability of
making Type I error, i.e. the level of significance of your test.
(b) Find the numerical value of c such that the test 𝑅: 𝑋̅ ≥ 𝑐 has a 5% of significance.

1 1
2. Suppose you want to test the hypotheses 𝐻0 : 𝑝 = 3 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝 ≠ 3 on the basis

of a random sample of size 10. If the rejection region is set as 𝑋 ≥ 5 , determine the
probability of making a type I error.

3. The probability that a person has an allergic reaction to a new drug on the market is 𝑝 =
0.28. The drug is modified and retested on 10 people. The null hypothesis 𝐻0 : 𝑝 = 0.28
is rejected if at most two people have an allergic reaction.
(a) What is the probability of making Type I error?
(b) Find the probability of making Type II error for 𝑝 = 0.20 and 𝑝 = 0.10.

Types of tests
One-tailed tests are those statistical tests with the rejection region either on the left or right
hand side of the distribution, i.e. those tests whose alternative hypothesis has the form
𝐻1 : 𝜇 > 𝜇0 or 𝐻1 : 𝜇 < 𝜇0 .

Two-tailed test are those statistical tests with rejection regions on either tails or ends of the
statistical distribution, i.e. those tests whose alternative hypothesis has the form 𝐻1 : 𝜇 = 𝜇0 .

The position of the rejection region depends on the form of the alternative hypothesis and its
size depends on the level of significance selected. The following diagram shows the
critical/rejection regions corresponding to each form of 𝐻1 .
Tests concerning the mean of a single population 𝝁

Consider the random sample 𝑋1 , 𝑋2 , ⋯ , 𝑋𝑛 form a normal distribution or population with

mean 𝜇 and variance 𝜎 2 . The test statistics that is used to test the following hypotheses

𝐻0 : 𝜇 = 𝜇0 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜇 > 𝜇0
𝐻0 : 𝜇 = 𝜇0 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜇 < 𝜇0
𝐻0 : 𝜇 = 𝜇0 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜇 ≠ 𝜇0

is given by
𝑥̅ −𝜇
(a) 𝑍 = 𝜎 ~𝑁(0, 1) if the population variance 𝜎 2 is known.
⁄ 𝑛
√
𝑥̅ −𝜇
(b) 𝑇 = 𝑠 ~𝑡(𝑛 − 1) if the population variance 𝜎 2 is unknown.
⁄ 𝑛
√
𝑥̅ −𝜇
(c) 𝑍 = 𝑠 ~𝑁(0, 1) if the population variance 𝜎 2 is unknown but the sample size is
⁄ 𝑛
√

large.

Example
Suppose we want to ascertain whether the mean amount of sulfur in mustard is 0.7, given that
from a sample of size 9, the mean is 0.706. It is known from past experience that the amount
of sulfur in mustard is normally distributed with a standard deviation of 0.25.

Solution

𝐻0 : 𝜇 = 0.7 𝑣𝑒𝑟𝑠𝑢𝑠 𝐻1 : 𝜇 ≠ 0.7

∝= 0.05
𝑥̅ −𝜇
Test statistic 𝑍 = 𝜎 ~𝑁(0, 1)
⁄ 𝑛
√

We reject 𝐻0 if |𝑍∝⁄2 | ≥ 1.96 (1.96 is obtained from the Normal distribution Tables)

Substituting 𝑥̅ = 0.706, 𝑛 = 9 and 𝜎 = 0.25 in the test statistic we have

0.706−0.7
𝑍= 0.25⁄ = 0.072.
√9
Since |𝑍| is less than 1.96, hence we fail to reject 𝐻0 , we conclude that there is sufficient
evidence that the mean sulfur content is no significantly different from 0.7.

Example

The following is a random sample of 9 observations on the profits (in $000) realised per month
by women cooperatives; 4.9, 5.8, 5.9, 6.5, 5.5, 5.0, 6.0, 5.6 and 5.7. Test at ∝= 0.05 whether
the mean profit is less than 5.9.

Solution

𝐻0 : 𝜇 = 5.9 𝑣𝑒𝑟𝑠𝑢𝑠 𝐻1 : 𝜇 < 5.9

∝= 0.05
𝑥−𝜇
Since the population variance is unknown the test statistic is 𝑡 = 𝑠⁄ ~𝑡(𝑛 − 1)
𝑛

We reject 𝐻0 if 𝑡 < −1.86 (from the t-distribution tables).

Substitute 𝑥̅ = 5.66 and 𝑠 = 0.49 from sample data into your test statistic, to obtain
5.66 − 5.9
𝑡= = −1.488
0.49⁄
√9
Since 𝑡 is greater than −1.86, we fail to reject 𝐻0 and conclude that at ∝= 0.05, the average
profit is less than $5900

Example
Suppose that the legally required level of sodium in the production of a certain beverage
is 3.5𝑔. A randomly selected sample of 144 bottles of the beverage gave a sample mean of
3.0𝑔 and a sample standard deviation of 2.7𝑔. Test whether that amount of sodium is lower
than 3.5𝑔. Use ∝= 0.05

Solution

𝐻0 : 𝜇 = 3.5 𝑣𝑒𝑟𝑠𝑢𝑠 𝐻1 : 𝜇 < 3.5

∝= 0.05
Since the population variance is unknown but the sample size is large, i.e. 𝑛 > 30 the test
𝑥̅ −𝜇
statistic is 𝑍 = 𝑠 ~𝑁(0, 1)
⁄ 𝑛
√

We reject 𝐻0 if 𝑍 ≥ 𝑍0.05 = −1.649 (from the Normal Tables)

Substitute 𝑥̅ = 3.0 and 𝑠 = 2.7 from sample data into your test statistic, we obtain
3.0−3.5
𝑍 = 2.7 = −2.22
⁄
√144

Since 𝑍 is less than -1.645, we reject 𝐻0 and conclude that at ∝= 0.05, we have enough
evidence that the mean sodium level in the beverages is less than 3.5g.

Exercises
1. For each of the following sets of values, test the hypotheses indicated.
(a) 𝑛 = 16, 𝑥̅ = 42, 𝑆 2 = 16: 𝐻0 : 𝜇 = 38 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜇 > 38
(b) 𝑛 = 9, 𝑥̅ = 5.9, 𝜎 2 = 6.25: 𝐻0 : 𝜇 = 4.9 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜇 ≠ 4.9
(c) 𝑛 = 49, 𝑥̅ = 21, 𝑆 2 = 2.89: 𝐻0 : 𝜇 = 18 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜇 < 18

2. The following data are amounts (in grams) of a certain vitamin found in raw milk.
7.97 7.83 7.56 6.15 7.99 7.28 8.11
8.09 7.12 6.69 7.54 7.35 4.55 6.77
Is there enough evidence to conclude that the mean amount of the vitamin is 6.00g? Use
∝= 0.05.

Tests Concerning the Difference between Two Population Means

The sampling distributions used in this section are the same as the ones derived in chapter **
and also ones used in the construction of confidence interval. In this section, we are to look at
situations where we are testing for equality of two population means or situations where the
difference between two population means is equal to some non-zero other value.

The following pairs of hypotheses will be tested.

𝐻𝑜 : 𝜇1 − 𝜇2 = 𝛿 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻0 : 𝜇1 − 𝜇2 > 𝛿
𝐻𝑜 : 𝜇1 − 𝜇2 = 𝛿 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻0 : 𝜇1 − 𝜇2 < 𝛿
𝐻𝑜 : 𝜇1 − 𝜇2 = 𝛿 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻0 : 𝜇1 − 𝜇2 ≠ 𝛿
If we are testing for the equality of two population means then, 𝛿 = 0, the hypotheses to be
tested then become
𝐻𝑜 : 𝜇1 = 𝜇2 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻0 : 𝜇1 > 𝜇2
𝐻𝑜 : 𝜇1 = 𝜇2 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻0 : 𝜇1 < 𝜇2
𝐻𝑜 : 𝜇1 = 𝜇2 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻0 : 𝜇1 ≠ 𝜇2

We will consider three different scenarios, given below:

(a) If population variances, 𝜎12 and 𝜎22 are both known

(𝑥̅ 1 −𝑥̅ 2 )−(𝜇1 −𝜇2 )

The test statistic is 𝑍 = ~𝑁(0, 1)
𝜎2 𝜎2
√( 1 + 2 )
𝑛1 𝑛2

(b) If population variances, 𝜎12 and 𝜎22 are both unknown and small sample sizes

(𝑥̅ 1 −𝑥̅ 2 )−(𝜇1 −𝜇2 ) 2 (𝑛1 −1)𝑆12 −(𝑛2 −1)𝑆22

The test statistic is 𝑇 = where 𝑆𝑝𝑜𝑜𝑙𝑒𝑑 = .
1 1 𝑛1 +𝑛2 −2
𝑆𝑝𝑜𝑜𝑙𝑒𝑑 √( + )
𝑛1 𝑛2

(c) If population variances, 𝜎12 and 𝜎22 are both unknown and sample sizes large, i.e. 𝑛1 >
30 and 𝑛2 > 30

(𝑥̅ 1 −𝑥̅ 2 )−(𝜇1 −𝜇2 )

The test statistic is 𝑍 = ~𝑁(0, 1)
𝑠2 𝑠2
√( 1 + 2 )
𝑛1 𝑛2

Example

Males claim that their weekly savings are higher than those of their female counterparts. It is
2 2
known that the variances for their weekly savings are 𝜎𝑓𝑒𝑚𝑎𝑙𝑒 = 7.84 and 𝜎𝑚𝑎𝑙𝑒 = 5.76. A
sample of 40 women gave an average saving of 5.22 whilst another sample of 37 men gave an
average of 6.45. Test at 𝛼 = 0.05 whether this claim can be justified.
Solution

𝐻𝑜 : 𝜇𝑚𝑎𝑙𝑒 = 𝜇𝑓𝑒𝑚𝑎𝑙𝑒 against 𝐻0 : 𝜇𝑚𝑎𝑙𝑒 > 𝜇𝑓𝑒𝑚𝑎𝑙𝑒

𝛼 = 0.05
(𝑥̅ 1 −𝑥̅ 2 )−(𝜇1 −𝜇2 )
The test statistic is 𝑍 = ~𝑁(0, 1)
𝜎2 𝜎2
√( 1 + 2 )
𝑛1 𝑛2

We reject 𝐻𝑜 if 𝑍 > 𝑍0.05 = 1.645 (from Normal distribution tables)

Substituting all symbols for their respective figures we get

(6.45 − 5.22) − (0)

𝑍= = 2.074
√(5.76 + 7.84)
37 40

We reject 𝐻𝑜 since 𝑍 > 1.645 and conclude that at 𝛼 = 0.05 males weekly savings of males
is greater than that of females.

Example
In the comparison of two food preservatives purchased by a certain food company, a random
sample of size 8 of one preservative gave an average shelf-life of 6.75 months with a variance
of 1.852 𝑚𝑜𝑛𝑡ℎ𝑠 2 while the another sample of size 13 of the other preservative gave an average
of 5.92 months with a variance of 1.442 𝑚𝑜𝑛𝑡ℎ𝑠 2 . Test if there is a significant difference
between these two preservatives. Use 𝛼 = 0.05.

Solution
𝐻𝑜 : 𝜇1 = 𝜇2 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻0 : 𝜇1 ≠ 𝜇2
𝛼 = 0.05
(𝑥̅ 1 −𝑥̅ 2 )−(𝜇1 −𝜇2 )
The test statistic is 𝑡 = 1 1
~𝑡(8 + 13 − 2) ≡ 𝑡(19)
𝑆𝑝𝑜𝑜𝑙𝑒𝑑 √( + )
𝑛1 𝑛2

We reject 𝐻𝑜 if |𝑡| ≥ 𝑡0.025 (19) = 2.09

2 (8−1)(1.852 )−(13−1)(1.442 )
𝑆𝑝𝑜𝑜𝑙𝑒𝑑 = = 2.570563259. Therefore 𝑆𝑝𝑜𝑜𝑙𝑒𝑑 = 1.6033
8+13−2
 (6.75 − 5.92) − (0)
𝑡= = 1.152
1 1
1.6033√(8 + 13)

Since |𝑡| < 2.09 we fail to reject and conclude that the shelf lives of the two preservatives are
not significantly different.

Example

Consider the following sample statistics.

Sample size Sample mean Sample variance

Sample 1 320 59.7 5.76
Sample 2 480 53.2 6.25

Test at ∝= 0.05 whether the difference between the two populations means is greater than
3.2.

Solution

𝐻𝑜 : 𝜇1 − 𝜇2 = 3.2 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜇1 − 𝜇2 > 3.2

∝= 0.05
(𝑥̅ 1 −𝑥̅ 2 )−(𝜇1 −𝜇2 )
The test statistic is 𝑍 = ~𝑁(0, 1)
𝑠 2𝑠 2
√( 1 + 2 )
𝑛1 𝑛2

We reject 𝐻𝑜 if 𝑍 > 𝑍0.05 = 1.6449.

Substituting in the test statistic we get

(59.7 − 53.2) − 3.2
𝑍= = 18.74
√(5.76 + 6.25)
320 480
Since 𝑍 > 1.6449 we reject 𝐻𝑜 and conclude that at ∝= 0.05, the difference between the
two populations means is greater than 3.2.
Exercises
1. Eight secretaries were taught using method 1 to type their documents and 6 others were
taught using method 2. Their typing speeds measured by words typed per minute are shown
in the table below.
Method 1: 41 34 38 42 44 39 36 40
Method 2: 56 45 50 54 48 47

Are the typing speeds significantly different for the two teaching methods? Test at the
5% level of significance.

2. An employment consultant asked selected workers from two different industries to fill in
a questionnaire on job satisfaction. The answers were scored 0 to 50 with higher scores
indicating greater job satisfaction. The data recorded are given in table below.

Industry 𝑛 𝑋̅ 𝑆
A 35 38.4 3.1
B 44 30.5 2.2

Is there a significant difference in the job satisfaction of the workers in the two industries?
Use ∝= 0.05.

Paired Samples
In the previous section we were testing the difference between two population means using
independent samples. In this section we are now considering situations were the samples are
not independent.

Suppose we want to decide on the basis of weights whether a certain diet has an effect of
reducing weight. In this experiment we measure the weights of the people concerned before
they are put on the diet and also measure their weights after the exercise. In this case we obtain
two samples, the before experiment sample and the after experiment sample, and we call these
samples paired/matched samples.
 Item 1 2 3 4 ⋯⋯⋯⋯⋯ 𝑛
Before the diet 𝑋1 𝑋2 𝑋3 𝑋4 ⋯⋯⋯⋯⋯ 𝑋𝑛

After the diet 𝑌1 𝑌2 𝑌3 𝑌4 ⋯⋯⋯⋯⋯ 𝑌𝑛

The 𝑋’𝑠 are representing weights before the diet programme and the 𝑌’𝑠 are weights of
individuals after the diet exercise. To handle this problem, one has to work with the differences
between the paired measurements, i.e. compute the differences 𝑑𝑖 = 𝑋𝑖 − 𝑌𝑖 , for 𝑖 =
1, 2, 3, ⋯ , 𝑛.

1
Using the new sample of the 𝑑𝑖 ′𝑠 one can compute the sample mean 𝑑̅ = 𝑛 ∑𝑛𝑖=1 𝑑𝑖 and the
2
1 (∑𝑛
𝑖=1 𝑑𝑖 )
sample variance 𝑠𝑑2 = 𝑛−1 (∑𝑛𝑖=1 𝑑𝑖2 − ).
𝑛

The pairs of hypotheses under consideration are

𝐻0 : 𝜇𝑑 = 0 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻0 : 𝜇𝑑 > 0
𝐻0 : 𝜇𝑑 = 0 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻0 : 𝜇𝑑 ≠ 0
𝐻0 : 𝜇𝑑 = 0 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻0 : 𝜇𝑑 < 0

The null hypothesis is equivalent to say the programme has no effect on reducing weight.
Interpreting the alternative hypothesis depends on how one defines the differences.

The test statistic for paired samples is given by

𝑑̅
𝑡= ~𝑡(𝑛 − 1)
𝑆𝑑
⁄
√𝑛
Example
Ten people were put on a special exercise programme for 8 weeks in order for them to lose
weight. The table below gives the weights (in kg) of the 10 people before and after the
programme.
Person 1 2 3 4 5 6 7 8 9 10
Before 90 98 89 110 104 78 69 82 66 59
After 87 98 85 105 103 77 62 76 64 55

Test at the ∝= 0.05 level of significance whether this programme reduces weight.
Solution

𝐻0 : 𝜇𝑑 = 0 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜇𝑑 > 0
∝= 0.05
Let 𝑑𝑖 = Before − After, for 𝑖 = 1, 2, ⋯ ,10 we obtain the sample
3, 0, 4, 5, 1, 1, 7, 6, 2, and 4
𝑑̅
The test statistic is given by 𝑡 = 𝑆𝑑 ~𝑡(9)
⁄
√𝑛

We reject 𝐻0 if 𝑡 ≥ 𝑡0.05 (9) = 1.83

Substituting 𝑛 = 10, 𝑑̅ = 3.3 and 𝑆𝑑 = 2.312 in the test statistic, we get
3.3
𝑡= = 4.514
2.312⁄
√10
Since 𝑡 > 1.83, we reject 𝐻0 and conclude that the programme indeed reduces weight.

Exercises
1. Two different methods of memorizing difficult material are being tested. Nine pairs of
students are matched according to their IQ and background and then assigned to one of the
two methods at random. A test is finally given to all the students and the results obtained
are as follows.
1 2 3 4 5 6 7 8 9
Method A 90 86 72 65 44 52 46 38 43
Method B 85 87 70 62 4 53 42 35 46

Using 5% level of significance, test whether there is a difference in the effectiveness of

the two methods.

2. An experiment was conducted to test the effect of continuous music on productivity. Ten
workers were selected at random and their productivity was measured one month without
music and another month with music. The table below gives the average number of item
produced per day.
 Worker 1 2 3 4 5 6 7 8 9 10
Without 8.37 5.01 7.24 6.35 6.24 4.73 7.82 5.67 8.01 6.98
music
With music 7.38 6.96 8.03 6.40 6.02 4.90 8.53 6.28 8.59 7.32

(a) Test the claim that there is no change in production.

(b) Construct a 98% confidence interval for the production difference.

Tests Concerning a Population Proportion

When testing a hypothesis on population proportion, the hypotheses to be tested are any of the
following pairs
𝐻0 : 𝑝 = 𝑝0 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝 > 𝑝0

𝐻0 : 𝑝 = 𝑝0 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝 ≠ 𝑝0

𝐻0 : 𝑝 = 𝑝0 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝 < 𝑝0

The test statistics of all the given pairs of hypotheses is

𝑝̂−𝑝0
𝑍= ~𝑁(0, 1).
√𝑝0 𝑞0⁄𝑛

where 𝑝̂ is the sample estimate of 𝑝.

Example

A manufacturer claims that at least 98% of his products care defect-free. A sample of 800 items
showed that 56 of them were defective. Test the claim at 1% level of significance.

Solution

𝐻0 : 𝑝 ≤ 0.02 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝1 > 0.02

∝= 0.01

𝑝̂−𝑝0
Test statistic is 𝑍 =
√𝑝0 𝑞0⁄𝑛
Reject 𝐻0 if 𝑍 ≥ 𝑍0.01 = 2.3263
56
−0.02
800
Substitute in the test statistic 𝑍 = 0.02×0.98
= 10.102
√
800

Since 𝑍 > 2.3263 we reject 𝐻0 and conclude that the manufacturer’s claim is false.

Exercises

1. A drug manufacturer claims that the new drug cures a certain disease in 80% of the cases.
In a random sample of 180 patients who used the drug, 105 of them found it to be effective.
Test the manufacturer’s claim at the 0.02 level of significance.

2. A random sample of 480 urban adolescents revealed that 96 of them were to vote
supporting a law supporting abortion. Similarly, 88 out of another random sample of 620
rural adolescents were in support of abortion. Test at 5% level of significance whether
there is no significant difference in the two proportions.

3. A dietician claimed the 42% of children in a certain district were undernourished. In order
to verify this claim a random sample of 800 children was taken and 240 of them were
found undernourished. Test the dietician’s claim at 3% level of significance.

Tests Concerning the Difference between Two Populations Proportions

Let 𝑝1 and 𝑝2 be two proportions of items or individuals having a characteristic of interest in

two populations 1 and 2 respectively. Independent samples are drawn from the two populations
𝑥 𝑥
and gave the statistics 𝑝̂1 = 𝑛1 and 𝑝̂2 = 𝑛2 .
1 2

The hypotheses to be tested when we are interested in the difference between two population
proportion is either one of the following,

𝐻0 : 𝑝1 = 𝑝2 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝1 > 𝑝2

𝐻0 : 𝑝1 = 𝑝2 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝1 ≠ 𝑝2

𝐻0 : 𝑝1 = 𝑝2 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝1 < 𝑝2

The test statistics is

 𝑝̂1 −𝑝̂2
𝑍= 1 1
~𝑁(0, 1)
√𝑝̂(1−𝑝̂)[𝑛 +𝑛 ]
1 2

𝑥 +𝑥
where 𝑝̂ = 𝑛1+𝑛2 the pooled estimate of the proportion.
1 2

Example
In order to test the effectiveness of a new anthrax vaccine, 240 infected cattle were given the
vaccine and 300 were not. All the 540 cattle were infected with anthrax. Among those which
were vaccinated, 60 of them died and among those which were not vaccinated 115 of them died
from the disease. Does vaccination reduces mortality rate? Use ∝= 5%.

Solution

Let the subscript 1 represents the vaccinated cattle and 2 the unvaccinated ones.

𝐻0 : 𝑝1 = 𝑝2 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝1 < 𝑝2

∝= 0.05.

𝑝̂1 −𝑝̂2
The test statistics is 𝑍 = ~𝑁(0, 1)
1 1
√𝑝̂(1−𝑝̂)[𝑛 +𝑛 ]
1 2

Reject 𝐻0 if 𝑍 ≤ 𝑍0.05 = −1.6449 (from normal tables)

1 23 60+115 35 73
𝑝̂1 = 4 , 𝑝̂2 = 60 and 𝑝̂ = 240+300 = 108 and 𝑞̂ = 108

1 23
−
4 60
𝑍= 35 73 1 1
= −3.290
√ ( )[ + ]
108 108 240 300

Since 𝑍 < −1.6449 we reject 𝐻0 and conclude that indeed the vaccine reduces mortality rate
at 5% level of significance.

Exercises
1. In a random sample of 400 males, 48 of them were found to have sexually transmitted
diseases. In another random sample of 630 females 111 of them were found to have
sexually transmitted diseases.
(a) Construct a 98% confidence interval of the difference in the proportions of those
with the sexually transmitted diseases.
 (b) Use your result in (a) to test the hypotheses
𝐻0 : 𝑝1 = 𝑝2 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝1 ≠ 𝑝2
2. In a study to determine the prevalence of poverty in rural and urban areas, a sample of 280
households was taken in urban areas and 99 of them were classified as poor. A sample of
600 households from the rural areas gave 148 of them were classified as poor. Test at the
2% level of significance whether poverty is high in the urban areas than in rural areas.

3. Test the following hypotheses for the given data.

(a) 𝐻0 : 𝑝1 − 𝑝2 = 0 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝1 − 𝑝2 < 0
𝑛1 = 250, 𝑥1 = 42, 𝑛2 = 400, 𝑥2 = 96, ∝= 0.05
(b) 𝐻0 : 𝑝1 − 𝑝2 = 0.25 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝1 − 𝑝2 < 0.25
𝑛1 = 200, 𝑥1 = 60, 𝑛2 = 500, 𝑥2 = 50, ∝= 0.02
(c) 𝐻0 : 𝑝1 − 𝑝2 = 0 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝑝1 − 𝑝2 ≠ 0
𝑛1 = 100, 𝑥1 = 50, 𝑛2 = 150 𝑥2 = 90, ∝= 0.01

Tests Concerning a Population Variance

From the sampling distributions of the sample variance, it was shown that

(𝑛−1)𝑆 2 2
𝜒2 = ~𝜒𝑛−1 .
𝜎2

When testing any of following pairs of hypotheses

>
𝐻0 : 𝜎 = 2
𝜎02 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝜎 ≠ 𝜎02
2

<

(𝑛−1)𝑆 2
The test statistic is 𝜒 2 = .
𝜎02

Example

A manufacturer of light bulbs claims that the average lifetime of his bulbs is 150 hours with a
standard deviation of 4.8 hours. A sample of 9 bulbs gave the following lifetimes (in hours)

142, 152, 143, 148, 157, 155, 160, 151, 158

Test whether the lifetimes of bulbs have a variance greater than 4.82 hours? Use 5%
significance level.
Solution

𝐻0 : 𝜎 2 = 4.82 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜎 2 > 4.82

∝= 0.05

(𝑛−1)𝑆 2
The test statistic is 𝜒 2 = ~𝜒82
𝜎02

Reject 𝐻0 if 𝜒 2 ≥ 𝜒82 (0.05) = 15.5

Substituting the following 𝑛 = 9, 𝑆 2 = 41.444 and 𝜎 2 = 4.82 we obtain

(9 − 1)(41.444)
𝜒2 = = 14.390
4.82

Since 𝜒 2 is less than 15.5 we reject 𝐻0 and conclude that the variance of the lifetime of the
bulbs is not significantly different from 4.82 at 5% significance level.

Exercises
1. The following data are amounts (in grams) of a certain vitamin found in raw milk.
7.97 7.83 7.56 6.15 7.99 7.28 8.11
8.09 7.12 6.69 7.54 7.35 4.55 6.77
Is there enough evidence to conclude that the variance of the vitamin is 1.2𝑔2? Use
∝= 0.05.

2. The following is a random sample of 9 observations on the profits (in $000) realised per
month by women cooperatives; 4.9, 5.8, 5.9, 6.5, 5.5, 5.0, 6.0, 5.6 and 5.7. Test at ∝= 0.05
whether the population variance is significantly different from 0.2.

Tests for Equality of Two Population Variance

There are times when we want to compare or test measures of variability. In order to test for
equality of two population variances we use the sampling distribution of the ratio of two sample
variance.

The hypotheses that will be under consideration are

𝐻0 : 𝜎12 = 𝜎22 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜎12 > 𝜎22

 𝐻0 : 𝜎12 = 𝜎22 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜎12 ≠ 𝜎22

𝐻0 : 𝜎12 = 𝜎22 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜎12 < 𝜎22

The hypotheses above can be written also as

𝜎12 𝜎12
𝐻0 : = 1 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : >1
𝜎22 𝜎22

𝜎12 𝜎12
𝐻0 : =1 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : ≠1
𝜎22 𝜎22

𝜎12 𝜎12
𝐻0 : 2 = 1 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 2 < 1
𝜎2 𝜎2

𝜎12 𝑆12
The test statistic for is 𝐹 = ~𝐹𝑛1−1,𝑛2−1 .
𝜎22 𝑆22

1
From our probability theory, it can be shown that 𝐹1−∝, 𝑣1,𝑣2 = 𝐹 . This is very important
∝, 𝑣2 ,𝑣1

in cases where the values of 𝐹1−∝ are not given in the tables.

Example
Consider the following two samples from two independent normal populations.

Sample 1: 8.2, 5.3, 6.5, 5.1, 9.7 and 10.8

Sample 2: 9.5, 8.3, 7.5, 10.9, 11.3, 9.3, 8.8 and 8.0

Determine whether there is sufficient evidence that 𝜎12 = 𝜎22 . Use ∝= 0.10.

Solution

𝜎12 𝜎12
𝐻0 : =1 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : ≠1
𝜎22 𝜎22

∝= 0.10.

𝑆2
The test statistic is 𝐹 = 𝑆12 ~𝐹5,7
2

1 1
Reject 𝐻0 if 𝐹 ≤ 𝐹0.95,5,7 = 𝐹 = 4.88 = 0.2049 or if 𝐹 ≥ 𝐹0.05,5,7 = 5.29
0.05,7,5
Substitute 𝑆12 = 5.552 and 𝑆22 = 1.8143 into the test statistic we get

5.552
𝐹= = 3.06
1.814

Since F is between 0.205 and 5.29, we do not reject 𝐻0 . We conclude that is not enough
evidence to conclude that the population variances are significantly different.

Exercises
1. Test the following hypotheses:
𝜎12 𝜎2
(a) 𝐻0 : = 1 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜎12 > 1
𝜎22 2

𝑛1 = 9, 𝑆12 = 265, 𝑛2 = 16, 𝑆22 = 198, ∝= 0.05

𝜎12 𝜎2
(b) 𝐻0 : = 1 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : 𝜎12 ≠ 1
𝜎22 2

𝑛1 = 11, 𝑆12 = 16.4, 𝑛2 = 7, 𝑆22 = 6.4, ∝= 0.05

2. Consider the following samples from two normal populations. Test whether the two
population variances are different. Use ∝= 0.05

Sample 1: 16 10 24 9 6 16 22
Sample 2: 27 17 10 32 9 15
3. Using data in question 1,
𝜎12
(a) Construct a 95% confidence interval for .
𝜎22

𝜎12 𝜎12
(b) Use your confidence interval (a) to test 𝐻0 : = 1 𝑎𝑔𝑎𝑖𝑛𝑠𝑡 𝐻1 : ≠1
𝜎22 𝜎22

4. The risk of an investment is at times measured by its variance on the return on investment.
In a comparison of the risk associated with two investments, monthly returns on two $1000
investments were recorded and the data is given in the table below.

Investment 1 15 9 28 -2 21 10 0 10 13 18

Investment 2 16 -2 -13 35 22 -18 36 -12

(a) Is the risk associated with investment 2 more than that of investment 1? Use ∝
= 0.05 .
(b) What assumptions have you put in place in order to answer part (a)?

𝜎12
(c) Construct a 98% confidence interval for .
𝜎22