CHAPTER 6: TEST OF HYPOTHESIS

6.1 Definitions
6.2 Procedures for Hypothesis Testing
6.3 Hypothesis Test for single mean and difference between 2 means
6.4 Hypothesis Test for Variance and Ratio Variance

Chapter 6

Test of Hypothesis

6.1

Definitions

Hypothesis Test:
It is a process of using sample data and statistical procedures to decide whether to reject or not to
reject a hypothesis (statement) about a population parameter value (or about its distribution
characteristics).
Null Hypothesis,

This is a statement in which a population parameter has a specific value. The null
hypothesis is so named because it is the starting point for the investigation. The null
hypothesis is initially assumed to be true. Therefore, it is the hypothesis to be tested.
Usually it takes the form of a statement about population parameter containing an equal
sign,= .

Alternative Hypothesis,

It is a statement about the same population parameter that is used in the null hypothesis
and generally this is a statement that specifies the population parameter has a value
different in some way, from the value given in the null hypothesis. The rejection of the
null hypothesis will imply the acceptance of this alternative hypothesis.

Usually it contains the mathematical symbols >, < or .

Example 6.1
State the null and alternative hypothesis for each claim given below
(a) The average time to read a certain passage is 15 minutes. An educator claimed that a course
in speed reading will shorten the reading time.

(b) A chemist said that he invested an additive which can increase the life of batter. The mean
lifetime is 24 months.
(c) The mean waiting bus for buses in Klang Valley is 8 minutes. Some roads are restricted to
buses only during office hours. A test is performed to see how this has affected the mean
waiting time.
Solution:

Test statistic:
It is a function of the sample data on which the decision is to be based.
Critical region:
It is a set of values of the test statistics for which the null hypothesis will be rejected.
Critical value (point):
It is the first (or boundary) value in the critical region.
Table 6.1 : Hypothesis and rejection region
One-tailed test
Left-tailed test
Hypothesis

Rejection region

Right-tailed test

Two-tailed test

Example 6.2
Determine whether the one-tailed test or two-tailed test is appropriate for the situation given
below:
(a) Testing whether the newly-purposed highway speed limit increases the number of accidents
(b) Testing whether the mean weight of chicken changed by breeding with other brand of
chicken feed
(c) A manufacturer of brake cables tests to see whether the breaking strength is increased with
implementing new technology in manufacturing process
(d) A manufacturer of ball-bearings tests to see whether the diameters are correct
Solution:

Example 6.3
Find critical value for each situation using standard normal distribution Z and draw the
appropriate figure, showing the rejection region.
(a) A right-tailed test with
(b) A left-tailed test with
(c) A two-tailed test with
Solution:

Whenever statistical inference methods are employed, it is always possible that the decision
reached will be incorrect. This is because partial information obtained from the sample is used to
draw conclusions about the entire population.
Type I and Type II Errors
There are two types of errors, namely Type I and Type II errors that may have to be considered
when performing the hypothesis testing.
Table 6.2 : Type I and II error
is true
Do not reject

Correct decision

Type II error

Reject

Type I error

Correct decision

Type I error : Rejecting

ie;

is false

when it is in fact true.

(level of significance) is the probability of making a Type I error.

Type II error : Accepting

when it is in fact false.

The probability of making such errors may be labeled as follows:

ie;

is the probability of making a Type II error.

6.2 Procedures for Hypothesis Testing

Step 1 : Define the question to be tested and formulate a hypothesis for stating the problem.
Step 2: Design the study.
-Choosing a level of significance.
-Selecting the suitable statistical test
-Finding the critical value(s) and determine the rejection region(s)
Step 3: Compute the test value
Step 4: Make the decision whether reject or not to reject the null hypothesis
Step 5: Summarize the results.

6.3

Hypothesis Test for Mean

To choose the appropriate statistical test for testing a claim about single population mean, we
follow the criteria below:
Choose Z or t-distribution ?

Is

known?

YES

Use Z distribution,

NO
YES

Use Z distribution,

NO

Use t-distribution,
Figure 6.1 : Criteria for choosing the appropriate statistical test
After selecting the test statistic and determine its critical value (s) as well as its test value, we
make the decision to rejct or not to reject the null hypothesis. If the test value falls in the
rejection region, we reject

. If not, we do not reject

, as shown before in Table 6.1.

6.3.1 Hypothesis Test for Single Mean

Testing the hypothesis about single population mean
Case 1: Variance,

known, for any value of sample sizes.

Statistical Test

Rejection Region

a)

a)

b)
c)

b)
c)

or

Case 2: Variance,

unknown and sample size

Statistical Test

Rejection Region

a)

a)

b)

b)

c)

Case 3: Variance,

c)

unknown and sample size

Statistical Test

Rejection Region

a)

a)

b)
c)

or

b)
c)

or

Example 6.4
A sport biologist claimed that female distance runners tend to be taller on the average than
women in general, who has an average height of 160 cm and standard deviation 8.75 cm. To
study this, she obtained a sample of 45 female distance runners and recorded their height, with
average height
Using these results, test the claim at 5% level of significance.

Example 6.5
The mean lifetime for a sample of 100 batteries is 1570 hours with standard deviation 120 hours.
However, the company claimed that their batteries average lifetime is 1600 hours. Test the claim
at 5% level of significance.

Example 6.6
A city health department wishes to determine whether the mean bacteria count per unit volume
of the water at a lake is within the safety level of 200. A researcher collected 10 water samples of
unit volume and found the bacteria counts to be
175

190

207

193

184

204

205

193

196

180

Do the data strongly indicate that the mean bacteria count is within the safety level? Test at 5%
level of significance.

6.3.2 Hypothesis Test for Difference between Two Means

To test a claim about difference between two population means, we follow the procedure
of solving hypothesis testing problems in subtopic 6.2. The statistical test (Step 2) for difference
between 2 means is given as below:
The Hypothesis:

Reject

when

Reject

when

Reject

when

or

or

The Test Statistics:

Hypothesis test involving two population mean,
Case 1: Variance,

and

can be categorized as follows:

known, for any value of sample sizes.

Case 2 and 3: Variance,

and

unknown, then the following table shows the different

formulaes that may be used depending on the sample sizes and the assumption on the population
variances.
Equality of

Sample sizes

variances

Case 2(a)

Case 3(a)

Case 2(b)

Case 3(b)

Example 6.7:
The mean lifetime of 30 bulbs produced by Company A is 50 hours and the mean lifetime of 35
bulbs produced by Company B is 48 hours. If the standard deviation of all bulbs produced by
Company A is 3 hours and the standard deviation of all bulbs produced by Company B is 3.5
hours, test at 1% significance level at the mean lifetime of bulbs produced by Company A is
better than that of Company B.

Example 6.8:
A sample of 60 maids from country A earns an average of RM 300 per week with a standard
deviation of RM 16, while a sample of 60 maids from country B earns an average of RM 250 per
week with a standard deviation of RM 18. Test at 5% significance level that country A maids
average earning exceed country B maids average earning more than RM 40 per week.

Example 6.9:
An experiment is done to test the strength of two types of glasses. A sample of 12 piecies of
glasses has a mean strength of 40 kg and a standard deviation of 2 kg. A sample of 13 piecees of
glasses has a mean strength of 38 kg and standard deviation of 2.5 kg. Test at 5% significance
level that the mean strength of the two types of glasses is the same. Assume the two population
variances are equal.

6.4 Hypothesis Test for Variance,

and Ratio Variance

6.4.1 Hypothesis Test for Variance,

The same procedure as in subtopic 6.2, but the chi-square distribution is used.
Testing Hypothesis about single population Variance
Statistical Test

Rejection Region

a)

a)

a)
a)
= specified value
Example 6.10:

b)
c)

A soft drink dispensing machine is said to be out of control if the variance of the contents
exceeds 1.15 deciliters. If a random sample of 25 sample of drinks from this machine has a
variance of 2.03 deciliters, does this indicate at the 0.05 level of significance that the machine
out of control? Assume that the contents are approximately normally distributed.

6.4.2 Hypothesis Test for Ratio of two Variance,

Testing Hypothesis about ratio of population Variances
Statistical Test

Rejection Region

a)
a)
a)

a)
or

or
a)
a)
a)

b)
c)

Example 6.11:
A machine cuts 12 rulers before it was serviced. The observed length for the 12 rulers has a
variance of 3.6
length is 2.5
level.

After the machine was serviced 9 rulers were cut and the variance of its
. Did servicing the machine improve its operation? Test at 5% significance