Chapter 5 : Hypothesis Testing

CHAPTER 5 : HYPOTHESIS TESTING

Sub-Topic
 Introduction to hypothesis testing.
 Terms in hypothesis testing.
 Type I and Type II errors.
 Type of hypothesis testing.
 Testing of hypothesis on a single population mean.
 Testing of hypothesis on a difference between two population means.
 Testing of hypothesis on a single population variance.
 Testing of hypothesis on a variance population ratios.

Chapter Learning Outcome

 Define the theory of statistics hypothesis and solve the problems involve
claims of mean and variance.

Learning Objective
By the end of this chapter, students should be able to
 Understand the basic of hypothesis testing.
 Identify the terms in hypothesis testing.
 Know the types of error.
 Know the types of hypothesis testing.
 Know the procedure to test a claim about single and different mean.
 Know the procedure to test a claim about single and ratio variance.

Key Term (English to Bahasa Melayu)

English Bahasa Melayu

1. Hypothesis testing → Pengujian hipotesis
2. Statistical inferences → Pentaabiran statistik
3. Null hypothesis → Hipotesis nol
4. Alternative hypothesis → Hipotesis alternatif
5. Type I error → Ralat jenis I
6. Type II error → Ralat jenis II
7. Significance level → Aras keyakinan

195
 Chapter 5 : Hypothesis Testing

5.1 Introduction to hypothesis testing

Hypothesis testing is also called significance testing. The objective of the procedure
is to test claims about parameters based on a random sample. A hypothesis test allows
us to draw conclusions or make decisions regarding population from sample data.
Statistical hypothesis testing is a decision-making process for evaluating claims about
a population. In hypothesis testing, the researcher must define the population under
study, state the particular hypotheses that will be investigated, give the significance
level, select a sample from the population, collect the data, perform the calculations
required for the statistical test and reach a conclusion. Researchers are interested in
answering many types of questions. For example such as “Will a new drug lower
blood pressure?” or “Will seat belts reduce the severity of injuries caused by
accidents?”. These types of questions can be addressed through statistical hypothesis
testing, which is a decision-making process for evaluating claims about a population.

5.2 Terms in hypothesis testing

In general, we do not know the true value of population parameters, they must be
estimated. However, we do have hypotheses about what the true values are. The
major purpose of hypothesis testing is to choose between two competing hypotheses
about the value of population parameter. For example, one hypothesis might claim
that the wages of men and women are equal, while the alternative might claim that
men make more than women.

Definition 1
A hypothesis is a statement about a population parameter.

Definition 2
The two complementary hypotheses in a hypothesis testing problems are called the
null hypothesis and the alternative hypothesis. They are denoted by H0 and H1,

196
 Chapter 5 : Hypothesis Testing

respectively. Both the null and alternative hypothesis should be stated before any
statistical test of significance is conducted. In other words, we technically are not
supposed to do the data analysis first and then decide on the hypotheses afterwards.

Definition 3
A hypothesis test is a rule that specifies for which sample values the decision is made
to reject H0 , i.e accept H1 and for which sample values not to reject H0.

Definition 4
The hypothesis actually to be tested is usually given the symbols H0, and is
commonly referred to as null hypothesis. The null hypothesis is assumed to be true
unless there is strong evidence to the contrary (similar to how a person is assumed to
be innocent until proven guilty). The null hypothesis always includes the equal sign,
which is H0 : μ = μ0.

Definition 5
The other hypothesis, which is assumed to be true when the null hypothesis is false, is
referred to as the alternative hypothesis. The alternative hypothesis always includes
three signs which is H1 : μ ≠ μ0, H1 : μ > μ0 and H1 : μ < μ0.

Definition 6
Test statistic is the sample statistic used to decide whether to reject or fail to reject the
null hypothesis.

Definition 7
Critical region is the set of all values which would cause us to reject H0.

Definition 8
Critical values are the values which separate the critical region from the non-critical
region. The critical values are determined independently of the sample statistics.

197
 Chapter 5 : Hypothesis Testing

Definition 9
Significance level (alpha) is the probability of rejecting the null hypothesis when it is
true. The value of α = 0.05 and α = 0.01 are common. If no level of significance is
given, use α = 0.05. The level of significance is the complement of the level of
confidence in estimation.

Definition 10
Decision is a statement based upon the null hypothesis. It is either reject the null
hypothesis or fail to reject the null hypothesis. We will never accept the null
hypothesis.

Definition 11
Conclusion is a statement which indicates the level of evidence (sufficient or
insufficient), at what level of significance and whether the original claim is rejected
(null) or supported (alternative).

5.3 Type I and Type II errors

Type I and Type II errors are two well-known concepts in quality engineering, which
are related to hypothesis testing. Often engineers are confused by these two concepts
simply because they have many different names. We list a few of them here.
Type I errors are also called :
 Producer’s risk.
 False alarm.
 False negative.
 α error.
Type II errors are also called :
 Consumer’s risk.
 Misdetection.
 False positive.

198
 Chapter 5 : Hypothesis Testing

 β error.
Type I and Type II errors can be defined in terms of hypothesis testing.
 A Type I error (α) is the probability of rejecting a true null hypothesis.
 A Type II error (β) is the probability of failing to reject a false null hypothesis.

Or simply :
 A Type I error (α) is the probability of telling you things are wrong, given that
things are correct.
 A Type II error (β) is the probability of telling you things are correct, given
that things are wrong.

The above statements are summarized in Table 1.

When H0 is true When H0 is false

Do not reject H0 No error Type II error
(Correct decision) p
p  1
Reject H0 Type I error No error
p  (Correct decision)
p  1 
Table 1 : Summary of Type I and Type II Errors

One concept related to Type II errors is "power." Power is the probability of rejecting
H0 when H1 is true. The value of power is equal to 1   . It is the power to detect the
change. The decision to reject or not to reject the null hypothesis is based on a test
statistic computed from values of a random sample. Hence, such a decision is subject
to error because of sampling variation. We denoted the probabilities of Type I and
Type II errors by

α = P(type I error) = P(reject H0 | H0 is true)

β = P(type II error) = P(not reject H0 | H0 is false)

199
 Chapter 5 : Hypothesis Testing

Definition 12
Type I error is rejecting the null hypothesis when it is true.

Definition 13
Type II error is failing to reject the null hypothesis when it is false.

Example 1
Indicate whether the following statements are type I or type II error.

(a) In a population, there is no relationship between alcohol consumption and

hours of sleep. In a sample of 5000 individuals, we accept the null hypothesis
that there is no relationship between alcohol and hours of sleep.

(b) In a population, there is a relationship between the number of times a person

vacuum a carpet and mental disorders. In a sample of 700 households, we
accept the null hypothesis that there is no relationship between carpet cleaning
and mental disorders.

Answer Example 1
(a) Type I error.
(b) Type II error.

Example 2
Assume that we are conducting a hypothesis test of the claim that μ < 0.06. Here are
the null and alternative hypotheses : H0 : μ = 0.06 and H1 : μ < 0.06.
Give the statements identifying
(a) Type I error.
(b) Type II error.

200
 Chapter 5 : Hypothesis Testing

Answer Example 2
(a) A type I error is the mistake of reject a true null hypothesis. Conclusion, there
is sufficient evidence to support μ < 0.06, when in reality μ = 0.06.
(b) A type II error is the mistake of fail to reject the null hypothesis when it is
false. Fail to reject μ = 0.06 (fail to support μ < 0.06) when in reality μ < 0.06.

5.4 Type of hypothesis testing

Tests of hypothesis can be carried out on one or two samples. One sample tests are
used to test if the population parameter (μ) is different from a specified value. Two
sample tests are used to detect the difference between the parameters of two
populations (μ1 and μ2). Two sample tests can further be classified as unpaired or
paired two sample tests. While in unpaired two sample tests the sample data are not
related, in paired two sample tests the sample data are paired according to some
identifiable characteristic. For example, when testing hypothesis about the effect of a
treatment on (say) a landfill, we would like to pair the data taken at different points
before and after implementation of the treatment.

One-tailed test
Here the alternate hypothesis H0 is one-sided and we test whether the test statistic
falls in the critical region on only one side of the distribution.

 One sample test

For example, we are measuring the concentration of a lake and we need to know
if the mean concentration of the lake is greater than a specified value of 10mg/L.
Hence, H0 : μ = 10 mg/L vs H1 : μ > 10 mg/L.

201
 Chapter 5 : Hypothesis Testing

 Two sample test

We want to test whether the population mean of the first sample is less than that
of the second sample.
Hence, H0 : μ1 = μ2 vs H1 : μ1 < μ2.

Example 3
In a manufacturing plant, plastic sheathing is specified to be at least two mils thick by
one of the many quality measures. Set up the null and alternative hypothesis for a
quality monitoring system that ensures the desired level of quality.

Answer Example 3
The machine operator would act by adjusting the extruder rollers on the machine only
if the plastic sheathing was too thin.
Null hypothesis, H0 : μ = μ0
Alternative hypothesis, H1 : μ < μ0
Where μ0 = 2 mils

Two-tailed test
Here the alternate hypothesis H1 is formulated to test for difference in either direction,
i.e., for either an increase or a decrease in the random variable. Hence the test statistic
is tested for occurrence within either of the two critical regions on the two extremes
of the distribution.
 One sample test
For the lake example we need to know if the mean concentration of the lake is
the same as or different from a specified value of 10 mg/L.
Hence, H0 : μ = 10 mg/L vs H1 : μ ≠ 10 mg/L.

202
 Chapter 5 : Hypothesis Testing

 Two sample test

We want to test whether the population mean of the first sample (μ1) is the
same as or different from the mean of the second sample (μ2).
Hence H0 : μ1 = μ2 vs H1 : μ1 ≠ μ2.

Example 4
In nuclear power plant, the cold start procedure consists if bringing the reactor to 35%
of power, and then to 65% of power, before full operation, a process that may take 12
hours. At each stage, engineers take measurements of several critical reactor
attributes. For examples, if binding energy for a given fuel rod does not have a mean
rate of 11.5MeV at 35% power, then the reactor could cascade into a critical
configuration and leak radiation at subsequent power levels. Set up the hypothesis for
a decision system at the 35% power level stage.

Answer Example 4
The plant operators would not continue to power up the reactor if the binding energy
did not meet specification. The action to be taken would be to shut down.
Null hypothesis, H0 : μ = μ0
Alternative hypothesis, H1 : μ ≠ μ0
Where μ0 = 11.5

Exercise 5.4
State the null and alternative hypotheses for each conjecture.

1. A researcher thinks that if pregnant women use vitamin pills, the mean birth
weight of the babies will increase. The average birth weights of the population
are 3.2 kilograms.

2. An engineer claim that the mean number of defects can be decreased in a

manufacturing process of compact discs by using robots instead of humans for

203
 Chapter 5 : Hypothesis Testing

certain tasks. The mean number of defective discs is eleven.

3. A psychologist feels that playing soft music during a test will change the
results of the test. The psychologist is not sure whether the grades will be
higher or lower. In the past, the mean of the scores was 95.

4. The mean waiting bus for travel to Gunung Tahan is 3.1 hours. Some roads
are restricted to buses only during office hours. A test is performed to see how
this bus has affected the mean waiting time.

5. The mean for incoming call received by Amir is 6 calls per hour. Amir
claimed that a call received will shorten the incoming calls.

6. A powder diet is tested on 49 people and a liquid diet is tested on 36 different

people. The population standard deviations are 2kg and 3kg, respectively. Of
interest is whether the liquid diet yields a higher average weight loss than the
powder diet.

7. Two types of chocolate bars are taste-tested on consumers. Of interest is

whether a type 1 number of customer have mean greater than type 2 number
of customer.

8. The average number of English courses taken in a two–year time period by

male and female college students is believed to be about the same. An
experiment is conducted and data are collected from 9 males and 16 females.

9. A football league reported that the average number of touchdowns per game
was 5. A study is done to determine if the average number of touchdowns has
decreased.

204
 Chapter 5 : Hypothesis Testing

10. A study is done to determine if students in the Universiti Kebangsaan

Malaysia (UKM) take longer to graduate than students enrolled in private
universities. 100 students from both the UKM and private universities are
surveyed. From years of research, it is known that the population standard
deviations are 1.5811 years and 1 year, respectively.

11. A recent drug survey showed an increase in use of drugs and alcohol among
local high school students as compared to the national percent. Suppose that a
survey of 100 local youths and 100 national youths is conducted to see if the
percentage of drug and alcohol use is higher locally than nationally.

12. University of Singapore researchers reported in the Journal of the National

Cancer Institute that quitting smoking is especially beneficial for those under
age 49. In this Singapore Cancer Society study, the risk (probability) of dying
of lung cancer was about the same as for those who had never smoked.

13. A golf instructor is interested in determining if new technique for improving

player 1 greater that player 2 golf scores is effective. She takes two new
students and records their 18-holes scores before learning the technique and
then after having taken her class.

14. Suppose a statistics instructor believes that there is no significant difference

between the average class scores of her two classes on Exam 2. The average
and standard deviation for her 8:30 class of 35 students were 75.86 and 16.91.
The average and standard deviation for her 11:30 class of 37 students were
75.41 and 19.73. “8:30” subscript refers to the 8:30 class. “11:30” subscript
refers to the 11:30 class.

15. A researcher is interested in determining if a certain drug vaccine prevents

some disease. The vaccine with the drug is administered to 36 people and

205
 Chapter 5 : Hypothesis Testing

another 36 people are given a vaccine that does not contain the drug. Conduct
a hypothesis test to determine if the person that get the vaccine without the
drug and get some disease is more than the people that get the vaccine with
the drug and get some disease.

Answer Exercise 5.4

1. H0 : μ = 3.2 vs H1 : μ > 3.2.
2. H0 : μ = 11 vs H1 : μ < 11.
3. H0 : μ = 95 vs H1 : μ ≠ 95.
4. H0 : μ = 3.1 vs H1 : μ ≠ 3.1.
5. H0 : μ = 6 vs H1 : μ < 6.
6. H0 : μP = μL vs H1 : μP < μL.
7. H0 : μ1 = μ2 vs H1 : μ1 > μ2.
8. H0 : μM = μF vs H1 : μM ≠ μF.
9. H0 : μ = 5 vs H1 : μ < 5.
10. H0 : μK = μP vs H1 : μK > μP.
11. H0 : μH = μN vs H1 : μH > μN.
12. H0 : μS = μN vs H1 : μS ≠ μN.
13. H0 : μ1 = μ2 vs H1 : μ1 > μ2.
14. H0 : μ1 = μ2 vs H1 : μ1 ≠ μ2.
15. H0 : μD = μN vs H1 : μD > μN.

5.5 Testing of hypothesis on a single population mean

__
In hypothesis single population mean, we are testing μ not testing x . If we knew the
value of μ, then there would be nothing to test.

Hypothesis testing steps

Here are the steps to performing hypothesis testing

206
 Chapter 5 : Hypothesis Testing

Step 1
Write the original claim and identify whether it is the null hypothesis or the
alternative hypothesis.
Step 2
Use the alternative hypothesis to identify the type of test.
Write down all information from the problem and with specific case.
State the distribution should use. Find the critical value using the tables and state the
decision rule.
Step 3
Compute the test statistic.
Step 4
Make a decision to reject or fail to reject the null hypothesis. A picture showing the
critical value and test statistic may be useful.
Step 5
Write the conclusion.

Population Standard Deviation (σ) Known

If the population standard deviation, σ, is known, then the population mean has a
normal distribution, and we will be using the Z-score formula for sample means. The
critical value is obtained from the normal table. With population standard deviation, σ
known, there are two cases with difference sample size such as below but same
statistics test.
_
x
Case A : n ≥ 30 with statistics test : Z Test  .
 n
_
x
Case B : n < 30 also with statistics test : Z Test  .
 n

Population Standard Deviation (σ) Unknown

If the population standard deviation, σ, is unknown with small sample size (n < 30),
then the population mean has a student's t distribution, and we will be using the t-

207
 Chapter 5 : Hypothesis Testing

score formula for sample means. The test statistic is very similar to that for the Z-
score, except that σ has been replaced by s and Z has been replaced by T. The critical
value is obtained from the t-table with the degrees of freedom for this test is n - 1.
Otherwise, if the population standard deviation, σ, also unknown with large sample
size (n ≥ 30), then the population mean has a normal distribution, and we will be
using the Z-score formula for sample means.
_
x
Case C : n ≥ 30 with statistics test : Z Test  .
s n
_
x
Case D : n < 30 with statistics test : TTest  .
s n

Example 5
A random sample of 120 recorded deaths in Filipina during the past years showed an
average life span of 71.8 years. Assuming a population standard deviation of 8.9
years, does this seem to indicate that the mean life span today is greater than 70 years
? Use a 0.01 level of significance.

Answer Example 5
Step 1
H0 : µ = 70 years
H1 : µ > 70 years (Claim)
Step 2
The right-tailed test. With n = 120, σ = 8.9 (known) and α = 0.01. This is Case A.
Use standard normal distribution. Critical value ZC > 2.33.
Decision Rule : Reject H0 if the test value, ZTest falls in the rejection region, ZC >
2.33.
Step 3

x  71.8  70
Z Test    2.22
 8.9
n 120

208
 Chapter 5 : Hypothesis Testing

Step 4
The test value is 2.22 which is less than the critical value, ZC > 2.33. The decision is
do not reject the null hypothesis.
Step 5
There is not enough evidence to support the claim that the mean life span today is
greater than 70 years.
Example 6
A researcher wishes to test the claim that the average age of lifeguards in Perhentian
Island is different than 33 years. He selects a sample of 14 guards and finds the mean
of the sample to be 32.1 years, with a sample standard deviation of 2 years. Is there
any evidence to support the claim by using alpha 0.05 ?

Answer Example 6
Step 1
H0 : µ = 33
H1 : µ ≠ 33 (Claim)
Step 2
The right-tailed test. With n = 14, s = 2 (σ unknown) and α = 0.05. This is Case D.
Use t-distribution. Critical value is tC < - 1.771 or tC > 1.771.
Decision Rule : Reject H0 if test value, Ttest falls in the rejection region, t C < - 1.771
or tC > 1.771.
Step 3
_
x   32.1  33
TTest    1.6837
s 2
n 14
Step 4
The test value is -1.6837 which is lower than the critical value, tC < - 1.771. The
decision is to do not reject the null hypothesis.
Step 5
There is not enough evidence to support the claim that the average age of lifeguards

209
 Chapter 5 : Hypothesis Testing

in Perhentian Island is different than 33 years.

Example 7
Random sample of 8 observations are taken to determine if there is evidence that the
concentration of an average certain material less than 11ppm. By using alpha equal
0.025, test the claim.
11.0 10.7 9.4 7.8
11.3 9.1 10.2 10.5

Answer Example 7
Step 1
H 0 :   11

H1 :   11 (Claim)
Step 2
The left-tailed test. With n = 8, σ unknown and α = 0.025. This is Case D.
 11  10.7  9.4  7.8  11.3  9.1  10.2  10.5
x
8
80
  10
8

11  102  10.7  102  9.4  102  7.8  102  11.3  102  9.1  102
 10.2  10  10.5  10
2 2
s
8 1

1  0.49  0.36  4.84  1.69  0.81  0.04  0.25


7

9.48

7
 1.1637
Use t-distribution. Critical value is tC < - 2.365.
Decision Rule : Reject H0 if test value, TTest falls in the rejection region, tC < - 2.365.

210
 Chapter 5 : Hypothesis Testing

Step 3
_
x   10  11
TTest    2.4305
s 1.1637
n 8
Step 4
The test value is - 2.4305 which is lower than the critical value, tC < - 2.365. The
decision is to reject the null hypothesis.
Step 5
There is enough evidence to support the claim that the concentration of an average
certain material less than 11ppm.

Example 8
The score of driving test has a normal distribution with mean 70 if given the standard
deviation of sample is eight. A driving school’s instructor claimed that if the
candidate learned more than three hours per week, the mean score would be different
than 70. A driving test was given to a random sample of 50 candidates with the mean
score was 78.
(a) State the null and alternative hypotheses.
(b) Identify the type I error and type II error that correspond to the hypothesis
above.
(c) Test the claim at 5% level of significance.

Answer Example 8
(a) H 0 :   70

H1 :   70 (Claim)
(b) Type I error : Reject H0 (The mean score is exactly 70).
Type II error : Do not reject H0 (The mean score is actually different than 70).
(c) Step 1
H 0 :   70

H 1 :   70 (Claim)

211
 Chapter 5 : Hypothesis Testing

Step 2
The right-tailed test. With   70 , s  8 , n  50 , x  78 and   0.05 . This
is Case C.
Use standard normal distribution. Critical value is ± ZC = ± 1.645.
Decision Rule : Reject H0 if ZTest falls in the rejection region, ZC < -1.645 or
ZC > 1.645.
Step 3
x   78  70
Z Test    7.07106
s 8
n 50
Step 4
The test value is 7.07106 which are greater than critical value, ZC > 1.645.
The decision is to reject null hypothesis.
Step 5
There is enough evidence to support the claim that the score would be
different than 70 if the candidate learned more than 3 hours per week.

Example 9
In the year 2004, the mean family size was 4.09. A sample of 22 families taken this
year by a researcher produced a mean family size of 5.01 with a population standard
deviation is 0.14. Using a 0.01 level of significance, test the hypothesis to claim that
the mean family size has decreased since 2004.

Answer Example 9
Step 1
H0 : µ = 4.09
H1 : µ < 4.09 (Claim)
Step 2
__
The left-tailed test. With n = 22, x  5.01 , σ = 0.14, α = 0.01. This is Case B.
Use standard normal distribution. Critical value is ZC < - 2.33.

212
 Chapter 5 : Hypothesis Testing

Decision Rule : Reject H0 if ZTest falls in the rejection region, ZC < - 2.33.
Step 3

x  5.01  4.09
Z Test    30.8227
 0.14
n 22

Step 4
The test value is 30.8227 which is lower greater than the critical value, ZC < - 2.33.
The decision is do not reject the null hypothesis.
Step 5
There is not enough evidence to support the claim that the mean family size has
decreased since 2004.

Exercise 5.5
1. A garment manufacturing company recorded the amount of time that it took to
make a pair of jeans on 8 different occasions. The time in minutes is as
follows.
12.5 13.0 11.9 10.2 13.1 13.6 13.8 14.0
Assume that the measurements were taken from the population with a normal
distribution. It is of interest to know if a sample data suggest that the average
time it takes this company to make a pair of jeans is less than 13.5 minutes.
State the null and alternative hypothesis, and then conduct an appropriate test
by using 0.05 of significance level.

2. A researcher wonders whether attending a private high school leads to higher

or lower performance on a test of social skills. The national means score for
students from public school is 75.62, if given the standard deviation is 28.0. A
sample of 100 students from a private school produces a mean score of 71.30
on the test. Using these results, test the claims at 5% level of significance.

213
 Chapter 5 : Hypothesis Testing

3. A researcher claims that the average salary of assistant professors is more than
RM42000. A sample of 30 assistant professors has mean salary of RM43260.
Test the claim at 5% level of significance that assistant professors earn more
than RM42000 a year. The standard deviation of the population is RM5230.

4. The score of driving test has a normal distribution with mean 70 and standard
deviation of population is 8. A driving school’s instructor claimed that if the
candidate learned more than three hours per week, the mean score would be
more than 70. A driving test was given to a random sample of 50 candidates
with the mean score was 78.
(a) State the null and alternative hypotheses.
(b) Identify the type I error and type II error that correspond to the
hypothesis above.
(c) Test the claim at 5% level of significance.

5. The researcher claim that the average cost of men’s athletic shoes is less than
RM80. He selects a random sample of 36 pairs of shoes from a catalog and
finds the following costs. Test the hypothesis 0.10 level of significance.
60 70 75 55 80 55 50 40 80 70 50 95 120 90 75 85 80 60
110 65 80 85 85 45 75 60 90 90 60 95 110 85 45 90 70 70

6. The Medical Rehabilitation Education Foundation claim that the average cost
of rehabilitation for stroke victims is RM24672. To see of the average cost of
rehabilitation is different at a particular hospital, a researcher selected a
random sample of 35 stroke victims at the hospital and found that the average
costs of their rehabilitation is RM25226. The standard deviation of the
population is RM3251. Test the hypothesis 0.01 level of significance.

7. A researcher wishes to test the claim that the average age of lifeguards in
Ocean City is greater than 24 years. She selects a sample of 36 guards and

214
 Chapter 5 : Hypothesis Testing

finds the mean of the sample to be 24.7 years, with a standard deviation
sample of 2 years. Test the hypothesis 0.05 level of significance.

8. A researcher claim that the average wind speed in a certain city difference
from 8 miles per hour. A sample of 32 days has an average wind speed of 8.2
miles per hour. The standard deviation of the sample is 0.6 mile per hour. Test
the hypothesis 0.05 level of significance.

9. The mean lifetime for a sample of 125 lamps is 1205 hours with standard
deviation 105 hours. However, the company claims that their lamps average
lifetime is difference from 1300 hours. Test the claim at 1% level of
significance.

10. A sample of 130 long-distance calls placed through a company produced a

mean length of 13.91 minutes with a standard deviation of 2.73 minutes. The
average length of all long-distance calls placed through this company in 2000
was 12.64 minutes. This company wanted to check if the mean length of the
current long-distance calls is different from 12.64 minutes. Using the 5%
significance level, test the hypothesis.

11. Drills being manufactured are supposed to have a mean length of 4cm. From
past experience, we know that the standard deviation is equal to 1cm and the
lengths are normally distributed. A random sample of 10 drills had a mean of
4.5cm. Test the hypothesis that the mean is 4.0 with α = 0.05.

12. An advertisement for a certain brand of cigarettes claimed that on average

there is no more than 18mg of nicotine per cigarettes. A test of 12 cigarettes
gave a sample mean of 19.1. Assuming variance is 4, test the claim with a
significance level of α = 0.05.

215
 Chapter 5 : Hypothesis Testing

13. A pharmaceutical manufacturer is concerned about the impurity concentration

in batches of drug and is anxious that the mean impurity doesn’t exceed 2.5%.
It is known that impurity concentration follows a normal distribution. A
random sample of 10 batches had the following concentrations.
2.1 1.9 2.4 2.3 2.6
1.5 2.8 2.6 2.7 1.8
Test at a significance level α = 0.05 that the population mean concentration is
at most 2.5.

14. Researchers studying the effects of diet on growth would like to know of a
vegetarian diet affects the height of a child. The researchers randomly selected
12 vegetarian children that are six years old. The average height of the
children is 42.5 inches with a standard deviation of 3.8 inches. The average
height for all six year old children is 45.75 inches. Conduct a hypothesis test
to determine whether there is overwhelming evidence at α = 0.05 that six year
old vegetarian children are not the same height as other six year old children.
Assume the heights of six year old vegetarian children are approximately
normally distributed.

15. In attempting to control the strength of the wastes discharged into a nearby
river, a paper firm has taken a number of measures. Members of the firm
believe that they have reduced the oxygen-consuming power of their wastes
from a previous mean of 500. If given sample size 25 with the variance of 0.9,
with significance level of α = 0.01. Test the claim if mean less than 550.

Answer Exercise 5.5

1. Do not reject H0.
2. Do not reject H0.
3. Do not reject H0.
4. (a) H 0 :   70 & H1 :   70 (Claim).

216
 Chapter 5 : Hypothesis Testing

(b) Type I error – Reject H0 when the mean score is equal 70.
Type II error – Do not reject H0 when the mean score is more than 70.
(c) Reject the H0.
5. Reject the H0.
6. Do not reject H0.
7. Reject the H0.
8. Do not reject H0.
9. Reject the H0.
10. Reject the H0.
11. Do not reject H0.
12. Reject H0.
13. Do not reject H0.
14. Reject H0.
15. Reject H0.

5.6 Testing of hypothesis on a difference between two means

There are 5 steps need to follow in order to solve hypothesis testing.

Hypothesis testing steps

Here are the steps to performing hypothesis testing
Step 1
Write the original claim and identify whether it is the null hypothesis or the
alternative hypothesis.
Step 2
Use the alternative hypothesis to identify the type of test.
Write down all information from the problem and with specific case.
State the distribution should use. Find the critical value using the tables and state the
decision rule.
Step 3

217
 Chapter 5 : Hypothesis Testing

Compute the test statistic.

Step 4
Make a decision to reject or fail to reject the null hypothesis. A picture showing the
critical value and test statistic may be useful.
Step 5
Write the conclusion.
Case Variances Samples size Statistical Test
A Known n1 , n2  30 ( X 1  X 2 )  ( 1   2 )
Z Test 
 12  22

n1 n2

B Known n1 , n2  30 ( X 1  X 2 )  ( 1   2 )
Z Test 
 12  22

n1 n2

C Unknown n1 , n2  30 ( X 1  X 2 )  ( 1   2 )
Z Test 
s12 s2
 2
n1 n2

D Unknown (Equal) n1 , n2  30 ( X 1  X 2 )  ( 1   2 )
TTest 
1 1
Sp  
n1 n 2

v  n1  n2  2
E Unknown (Not equal) n1  n 2  30 ( X 1  X 2 )  ( 1   2 )
TTest 

1 2
n
s1  s 22 

v  2(n  1)
F Unknown (Not equal) n1 , n2  30 ( X 1  X 2 )  ( 1  2 )
TTest 
s12 s2
 2
n1 n2

218
 Chapter 5 : Hypothesis Testing

2
 s12 s22 
  
v   12
n n2 
2
 s1 
2
 s22 
   
 1    n2 
n
n1  1 n2  1

Example 10
A sample of 35 teachers from Kedah has an average salary of RM32100, with a
standard deviation of RM1220. A sample of 32 teachers from Johor has an average
salary of RM31290, with a standard deviation of RM1320. Using alpha 0.01, perform
the hypothesis testing whether any significant difference in teachers’ salaries between
the two places. Assume the salaries are normally distributed.

Answer Example 10
Step 1
H0 : 1  2
H1 : 1  2 (Claim)

Step 2
Kedah Johor
Sample size 35 32
Sample mean 32100 31290
Sample standard deviation 1220 1320

The two-tailed test. With α = 0.05 and the data above. This is Case C.
Use standard normal distribution. Critical value is ZC < -2.58 or ZC > 2.58.

219
 Chapter 5 : Hypothesis Testing

Decision Rule : Reject H0 if the ZTest falls in the rejection regions, ZC < -2.58 or ZC >
2.58.
Step 3
(32100  31290)  (0)
Z Test   2.60107
(1220) 2 (1320) 2

35 32
Step 4
The test value is 2.60107 which are greater than the critical value, ZC > 2.58. The
decision is to do not reject the null hypothesis.
Step 5
There is not enough evidence to support the claim that there is significant difference
in teachers’ salaries between the two places.

Example 11
The music was turned on during the working hours of a business with 45
employees. There productivity level averaged 5.2 with a standard deviation of
2.4. On a different day, the music was turned off and there were 40 workers. The
workers' productivity level averaged 4.8 with a standard deviation of 1.2. Test the
claim whether employees with music playing perform better than employees without
music playing using 0.05 level of significance.

Answer Example 11
Step 1
H0 : 1  2

H1 : 1  2 (Claim)
Step 2
Turned On Turned Off
Sample size 45 40
Sample mean 5.2 4.8
Sample standard deviation 2.4 1.2

220
 Chapter 5 : Hypothesis Testing

The right-tailed test. With α = 0.05 and data above. This is Case A.
Use standard normal distribution. Critical value is ZC > 1.645.
Decision Rule : Reject H0 if the test value, ZTest falls in the rejection regions,
ZC > 1.645.
Step 3
(5.2  4.8)  (0)
Z Test   0.9877
(2.4) 2 (1.2) 2

45 40
Step 4
The test value is 0.9877 which is less than the critical value, ZC > 1.645. The decision
is to do not reject the null hypothesis.
Step 5
There is not enough evidence to support the claim that employees with music playing
perform better than employees without music playing.

Example 13
In Mathematics quiz, the sample sizes of two parts are 10 students. For Part I, the
mean score was 35 with standard deviation of 2.5, while in Part II, the mean score
was 24 with standard deviation of 2.1. Test the difference between the performances
of the two parts using 0.01 of significance level. Assume that the variances of
population are unknown but not equal.

Answer Example 13
Step 1
H 0 :  I   II

H1 :  I   II (Claim)
Step 2
Part I Part II
Sample size 10 10
Sample mean 35 24

221
 Chapter 5 : Hypothesis Testing

Sample standard deviation 2.5 2.1

The two-tailed test. With α = 0.01 and data above. This is Case E.
We use t-distribution. Critical value is TC < - 2.878 or TC > 2.878.
Decision Rule : Reject H0 if the test value, Ttest falls in the rejection region, TC < -
2.878 or TC > 2.878.

Step 3

TTest 
35  24   0  10.65403
1
10
2.52  2.12 
Step 4
The test value is 10.65403 which are greater than the critical value, TC > 2.878. The
decision is to reject the null hypothesis.
Step 5
There is enough evidence to support the claim that there is a significant difference
between the performances of two parts.

Example 14
The data survey credit card holders in Malaysia such as below.
Year 2005 2006
Sample mean 756 784
Sample size 14 24
Sample standard deviation 12 9

Test at 0.1 significance level of the mean credit card holders in 2005 and 2006 were
different. Assume that variances of population are unknown but not equal.

222
 Chapter 5 : Hypothesis Testing

Answer Example 14
Step 1
H0 :  A   B

H1 :  A   B (Claim)
Step 2
Year 2005 2006
Sample mean 756 784
Sample size 14 24
Sample standard deviation 12 9
The two-tailed test. With α = 0.1 and data above. This is Case F.
Use t-distribution. Critical value are TC < - 1.717 or TC > 1.717.
2
 12 2 9 2 
  
v  2
14 24  186.61511
2
  21.6155  22
 12 2   92  8.63338
   
 14    24 
14  1 24  1
Decision Rule : Reject H0 if the test value, TTest falls in the rejection region, TC < -
1.717 or TC > 1.717.
Step 3

TTest 
756  784  0  7.57567
12 2 92

14 24
Step 4
The test value is -7.57567 which is less than the critical value, TC < - 1.717. The
decision is to reject the null hypothesis.
Step 5
There is enough evidence to support the claim that there is a significant difference
between the mean credit card holders in 2005 and 2006.

223
 Chapter 5 : Hypothesis Testing

Example 15
The data is about the average of mileage record by two type of engine in Toyota
company. The sample size of engine type I is 18 with sample mean 114. While the
sample size of engine type II is 14 with sample mean 123. If sample standard
deviation both of engines are 1.6 and 1.7 respectively, test the hypothesis use 0.025
level of significance the average of mileage engine type I is lower than the average of
mileage engine type II. Assume that the variances population unknown but equal.

Answer Example 15
Step 1
H0 : I  II
H1 : I  II (Claim)
Step 2
The left-tailed test. With α = 0.025. This is Case D.
Use t-distribution. Critical value is TC < - 2.042.

S p2 
18  11.62  14  11.72  2.703
18  14  2
S p  1.64408

Decision Rule : Reject H0 if the test value, TTest falls in the rejection region,
TC < - 2.042.
Step 3
(114  123)  (0)
TTest   15.36189
1 1
1.64408 
18 14
Step 4
The test value is -15.36189 which is less than the critical value, TC < - 2.042. The
decision is to reject the null hypothesis.
Step 5
There is enough evidence to support the claim that the average gas mileage for engine

224
 Chapter 5 : Hypothesis Testing

type I is significantly less than the average gas mileage for engine type II.

Exercise 5.6
1. A sample of 32 teachers from Langkawi Island has an average salary of
RM2310 per month, with a standard deviation of RM122. A sample of 36
teachers from Tioman Island has an average salary of RM2612 per month,
with a standard deviation of RM111. Test the hypothesis if there are
significant differences in teachers’ salaries between the two islands use 0.05
of significance level.

2. Two types of drugs were used on 5 and 7 patients for reducing their weights in
Jerry’s 'slim-beauty' health club. Drug A was allopathic and drug B was
Herbal. The decrease in the weight after using drugs for six months was as
follows.
Drug A : 10 12 13 11 14
Drug B : 8 9 12 14 15 10 9
Test the hypothesis if there are significant differences in drug B and drug A by
using 0.001 of significance level. Assume that the variances of population are
unknown but equal.

3. The average annual cost of car insurance in 2004 for residents of Kuala
Lumpur was RM891, while for residents of Pulau Pinang was RM789. If
given that the sample size of both states is 14 with standard deviation of
sample, 3 and 6 respectively. Test the hypothesis if mean annual cost of car
insurance Kuala Lumpur is greater than the mean annual cost of car insurance
Pulau Pinang. Use 0.10 of significance level. Assume that the variances of
population are unknown but not equal.

225
 Chapter 5 : Hypothesis Testing

4. Two types of batteries are tested for their length of life and following results
are obtained. Is there a significant difference in the two batteries ? Test the
claim using 0.05 of significance level.

Battery A Battery B
Sample size 12 10
Sample mean 500 560
Variance population 100 121

5. To test the effect of a fertilizer on rice production, 24 equal plots of a certain

land are selected. Half of them were treated with fertilizer leaving the rest
untreated. Other conditions were the same. The mean production of rice on
untreated plots was 4.8 quintals with standard deviation of 0.4 quintal, while
the mean yield on the treated plots was 5.1 quintals with a standard deviation
of 0.36 quintal. Can we say that there is significant improvement in the
production of rice due to use of fertilizer at 0.05 level of significance ?
Assume that the variances of population are unknown but not equal.

6. An insurance company wants the hypothesis that the mean amount of

insurance held by professional men difference that held by professional
women. Accordingly, two independent simple random samples are taken from
appropriate professional listings of men and women. A sample of 200 men
reveals a mean amount of RM140000 with a standard deviation RM26000. A
sample of 400 women shows a mean amount of RM128000 with standard
deviation of RM3000. State the appropriate hypothesis and conduct a
statistical for them using the significance level 0.05.

7. A study is conducted to assess the difference in performance during the first

years of services between employees that stayed in a certain company during

226
 Chapter 5 : Hypothesis Testing

15 years and those who left the company. The performance is measured by the
company’s annual performance appraisals which produce ratings on a 5 point
scale, 1 for low performance and 5 for high performance. The data are
summarized in the table. Use α = 0.05 to test the hypothesis.
Stayers Leavers
Sample size 174 355
Sample mean 3.51 3.24
Sample standard deviation 0.51 0.52

8. To investigate the effect of amphetamines on water consumption, 15 lab rats

were injected with amphetamine and 10 with saline solution. The water
consumed by each rat in ml/kg of body weight was recorded and the results
are summarized in the following table.
Amphetamine Saline
Sample size 15 10
Sample mean 115 135
Sample standard deviation 40 15
Test the hypothesis between two difference means using α = 0.05.

Answer Exercise 5.6

1. Reject the H0.
2. Do not reject the H0.
3. Reject the H0.
4. Do not reject the H0.
5. Do not reject the H0.
6. Reject the H0.
7. Reject the H0.
8. Reject the H0.

227
 Chapter 5 : Hypothesis Testing

5.7 Testing of hypothesis on a single variance

The test of a single variance is performed using a chi-square test and the chi-square
distribution. Let X1, … , Xn be a random sample from a population which is N(μ ,σ 2)
where μ and σ2 are unknown. We consider now how to test a hypothesis about the
population variance, σ2. We shall present the results without justification. To test
H0 : σ2 = σ20 versus H1 : σ2 ≠ σ20 we use the test statistic such as below.
(n  1)  s 2
 2T  ~  2 n 1
 2
0

with the degree of freedom is n – 1 and always assumed that H0 is true. Conditions for
testing are
 The population has a normal distribution.
 The data is from a random sample.
 The observations must be independent of each other.
Testing is done in the same manner as before. Remember, all hypothesis testing is
done under the assumption the H0 true.

Example 16
A manufacturer of car batteries claims that the life of his batteries is approximately
normally distributed with a standard deviation equal to 0.7 year. If a random sample
of 15 of these batteries has a standard deviation of 0.5 years, test the hypothesis of
variance population greater than 0.49 year by using 0.01 of significance level.

Answer Example 16
Step 1
H0 : σ2 = 0.49
H1 : σ2 > 0.49 (Claim)
Step 2
Use α = 0.01. Critical value is  2 C   2 0.01,14  29.141 .
Use chi-square distribution.

228
 Chapter 5 : Hypothesis Testing

Decision Rule : Reject H0 if the test value,  2Test falls in the rejection region.
Step 3
(n  1) s 2 (15  1)(0.5) 2
 2 Test    7.14285 .
2 (0.7) 2
Step 4
The test value is 7.14285 which is less than the critical value,  2 C  29.141 . The
decision is to reject the null hypothesis.
Step 5
There is enough evidence to support the claim that the life of his batteries with
variance greater than 0.49 year.

Example 17
An extra preparation class is advertised to improve the scores with random sample of
30 data and a standard deviation is 2.1 hours which is approximately normally
distributed. Assume the standard deviation of the scores is 1.7 hours. Use alpha equal
0.025, test the hypothesis.

Answer Example 17
Step 1
H0 : σ2 = 2.89
H1 : σ2 > 2.89 (Claim)
Step 2
Give α = 0.025. The test is a Chi-Square test. We use chi-square distribution.
The critical value is  2C   2 , v   2 0.025, 29  45.722 .

Decision Rule : Reject H 0 if the test value,  2Test falls in the rejection region.

Step 3
(n  1) s 2 (30  1)(2.1) 2
 2
Test    44.2525
2 (1.7) 2

229
 Chapter 5 : Hypothesis Testing

Step 4
The test value is 44.2525 which is less than the critical value,  2 C  45.722 which is
not in the critical region. The decision is do not reject the null hypothesis.
Step 5
There is not enough evidence to support the claim that an extra preparation class is
advertised to improve the scores.

Exercise 5.7
1. The score of driving test has a standard deviation of sample 8. A driving
school’s instructor claimed that if the candidate learned not more than three
hours per week, the standard deviation score would be less than 16. A driving
test was given to a random sample of 51 candidates. Use alpha equal 0.01, test
the hypothesis.

2. A researcher wishes to test the claim that the standard deviation of lifeguards
in Sipadan Island is difference than 3.4 years. He selects a sample of 16
guards and finds the standard deviation of 6 years. Is there any evidence to
support the claim by using alpha 0.01 ?

3. A random sample of 23 recorded deaths in Filipina during the past years

showed a standard deviation of 8.9 years, does this seem to indicate that the
standard deviation life span today is greater than 7.1 years ? Use a 0.05 level
of significance.

4. A researcher wonders to know whether attending a private high school leads

to higher performance on a test of social skills whether standard deviation
score for students from public school greater than 28.1. A sample of 25
students from a private school produces a standard deviation of 23.6 on the
test. Using these results, test the claims 0.025 of significance level.

230
 Chapter 5 : Hypothesis Testing

5. By using 0.01 significant levels, test the claim that variance of women
supermodels weight is less than the variance of women weights in general.
The population standard deviation of the weights is 29 pounds. The weights
(in pounds) of nine randomly selected supermodels are shown in below.

125 119 115 119 127 128 105 128 123

6. It is important that the variance of the percentage impurity levels of a

chemical don’t exceed 4.0. A random sample of 20 consignments had a
sample variance of 5.62. Test the hypothesis that the population variance is at
most 4.0 at a 5% level of significance.

7. Suppose a sample 30 observations is drawn from a population with variance

population is 4.55 and the sample variance is 6.7. Test the hypothesis that the
sample comes from a population with a variance greater than 4.55 at a 5%
level of significance.

8. With individual lines at its various windows, a post office finds that the
standard deviation for normally distributed waiting times for customers on
Friday afternoon is 7.2 minutes. The post office experiments with a single
main waiting line and find that for a random sample of 25 customers, the
waiting times for customers have a standard deviation of 3.5 minutes. With a
significance level of 5%, test the claim that a single line causes lower
variation among waiting times (shorter waiting times) for customers.

9. An aircraft manufacturer is concerned about variability in the diameters of lids

used to seal the fuel tanks that are located inside aircraft wings. Only a narrow
range of diameters is acceptable. Lids that fit too tightly prevent air from
entering the tanks as the fuel is being used, creating a vacuum and ultimately
causing collapse of the wing structure. Lids that fit too loosely can allow fuel

231
 Chapter 5 : Hypothesis Testing

to be sucked out of the tank during flight, which is equally undesirable from
the point of view of flight safety. A test at the 2% significance level is to be
conducted with a random sample of 20 fuel-tank lids to see whether the
population variance of lid diameters equals 0.0001 inches squared, as
specified by engineers.

10. A laser machine tool is supposed to cut watch gears in precise thickness
averaging 500 microns with standard deviation of 4 microns. We take a
random sample of 10 watch gears and they have these thicknesses in microns.

500 490 510 501 499 502 497 503 500 499

Use the sample to test the claim that the tool cuts gears with a thickness
variance of 4 microns. Use a significance level of 0.01.

11. For randomly selected adults IQ scores are normally distribution with a mean
of 100 and standard deviation of 15. A sample of 24 randomly selected
college professors resulted in IQ scores having a standard deviation of 10.
Test the claim that the IQ scores for college professors is the same as the
general population that is 15. Use a 0.05 level of significance.

12. Tests in Mr. Wildmans past statistics classes have scores with a standard
deviation equal to 14.1. One of his current classes now has 27 test scores with
a standard deviation of 9.3. Use a 0.01 level of significance to test the claim
that this current class has less variation than past classes.

Answer Exercise 6.7

1. Do not reject the H0.
2. Reject the H0.
3. Reject the H0.

232
 Chapter 5 : Hypothesis Testing

4. Do not reject the H0.

5. Reject the H0.
6. Do not reject the H0.
7. Reject the H0.
8 Reject the H0.
11. Reject the H0.
12. Reject the H0.

5.8 Testing of hypothesis on a variance ratios

Introduction
The F-distribution is formed by the ratio of two independent chi-square variables
divided by their respective degrees of freedom. Since F is formed by chi-square,
many of the chi-square properties carry over to the F distribution such as below.
 The F-values are all non-negative.
 The distribution is non-symmetric.
 The mean is approximately 1.
 There are two independent degrees of freedom, one for the numerator and the
other one for the denominator.
 There are many different F distributions, one for each pair degrees of freedom.
The F-test is designed to test if two population variances are equal. It does this by
comparing the ratio of two variances. So, if the variances are equal, the ratio of the
variances will be one. All hypothesis testing is done under the assumption the null
hypothesis is true. If the null hypothesis is true, then the F test-statistic which is
F  S 21 S 22 can be simplified (dramatically). This ratio of sample variances will be
test statistic used. If the null hypothesis is false, then, we will reject the null
hypothesis that the ratio was equal to 1 and our assumption that they were equal. The
F test statistic is simply the ratio of two sample variances. There are several different
F-tables. Each one has a different level of significance. So, find the correct level of
significance first, and then look up the numerator degrees of freedom and the

233
 Chapter 5 : Hypothesis Testing

denominator degrees of freedom to find the critical value. We will notice that all of
the tables only give level of significance for right tail tests. Because the F distribution
is not symmetric, and there are no negative values, we may not simply take the
opposite of the right critical value to find the left critical value. The way to find a left
critical value is to reverse the degrees of freedom, look up the right critical value, and
then take the reciprocal of this value.

Avoiding Left Critical Values

Since the left critical values are a pain to calculate, they are often avoided altogether.
We can force the F-test into a right tail test by placing the sample with the large
variance in the numerator and the smaller variance in the denominator. It doesn’t
matter which sample has the larger sample size, only which sample has the larger
variance. The numerator degrees of freedom will be the degrees of freedom for
whichever sample has the larger variance (since it is in the numerator) and the
denominator degrees of freedom will be the degrees of freedom for whichever sample
has the smaller variance (since it is in the denominator). If a two-tail test is being
conducted, we still have to divide alpha by 2, but we only look up and compare the
right critical value.

Assumptions
 The larger variance should always be placed in the numerator.
 The test statistic is F = s12 / s22 where s12 > s22.
 Divide alpha by 2 for a two tail test and then find the right critical value.
 If standard deviations are given instead of variances, they must be squared.
 When the degrees of freedom aren't given in the table, go with the value with
the larger critical value (this happens to be the smaller degrees of freedom).
This is so that you are less likely to reject in error (type I error).
 The populations from which the samples were obtained must be normal.
 The samples must be independent.

234
 Chapter 5 : Hypothesis Testing

Example 18
An experiment was performed to compare the abrasive wear of two different
laminated materials. Eleven pieces of material 1 were tested by expose each piece to a
machine measuring wear. Ten pieces of material 2 were similarly tested. In each case,
the depth of wear was observed. The samples of material 1 gave an average (coded)
wear of 85 units with a sample standard deviation of 4, while the samples of material
2 gave an average of 81 and a sample standard deviation of 5. Use a 0.1 level of
significance to test the ratio of two populations.

Answer Example 18
Step 1
H 0 :  21   2 2

H1 :  21   2 2 (Claim)
Step 2
Material 1 Material 2
Sample mean 85 81
Sample size 11 10
Sample standard deviation 4 5

With α = 0.1. The d.f : v1  n1  1  11  1  10 and v2  n2  1  10  1  9 .

Critical values : FC  F0.05 (10, 9)  3.14 and

1 1
F0.95 (10, 9)    0.3306
F0.05 (9, 10) 3.025
We use F-distribution.
Decision Rule : Reject H0 if the test value falls in the rejection region, FC  0.3306

or FC  3.14 .

Step 3
s 21 4 2
FTest    0.64 .
s 2 2 55

235
 Chapter 5 : Hypothesis Testing

Step 4
The test value is 0.64 which is between both of critical value, FC  0.3306 and

FC  3.14 . The decision is do not reject the null hypothesis.

Step 5
There is not enough evidence to support the claim that an experiment was performed
to compare the abrasive wear of two different laminated materials.

Exercise 5.8
1. In a study of the effects supplement use, brand A and brand B users of
supplement in college were tested for body fitness, with the results given
below. Use a 0.1 significance level to test the claim that the population of
brand B supplement users has a variance different from that brand A users.
Brand A Brand B
Sample size = 25 Sample size = 13
Standard deviation = 2.4 Standard deviation = 2.1

2. A sample of 13 teachers from Terengganu has an average salary per month of

RM2500, with a standard deviation of RM115. A sample of 13 teachers from
Tanjung Karang has an average salary per month of RM2710, with a standard
deviation of RM98. Test the ratio of teachers’ salaries between Terengganu
and Tanjung Karang whether different or not by using 0.05 level of
significance.
3. A mathematics test is given in two classrooms. In the first classrooms with 25
students, the mean was 84.3 and the sample variance was 16.8. In the second
classrooms with 13 students, the mean was 83.7 and the sample variance was
42.6. Test the hypothesis whether the two classroom variances different or not
by using α = 0.05.

4. To investigate the effect of amphetamines on water consumption, 15 lab rats

were injected with amphetamine and 10 with saline solution. The water

236
 Chapter 5 : Hypothesis Testing

consumed by each rat in ml/kg of body weight was recorded and the results
are summarized in the following table.
Amphetamine Saline
n 15 10
__ 115 135
x
s 40 15

To compare the two variances, test the hypothesis using α = 0.05.

5. When comparing two normal populations, we can be interested in differences

between the variances of them and not just the means. The data in the table
correspond to the amount of ethylene oxide (ETO) (in mmgrms) measured in
the bloodstream of 30 subjects manipulating sterilizing hospital supplies. 11
subjects where randomly assigned to one task and 19 to another. Is there
enough evidence in the data to conclude that there are differences in the
variability of the ETO levels between the two groups ?
Task 1 Task 2
Sample size 11 19
Mean 5.60 5.90
Standard deviation 4.10 1.93

6. The variability in the amount of impurities present in a batch of chemicals

used for a particular process depends on the length of time that the process is
in operation. Suppose a sample of size 25 is drawn from the normal process
which is to be compared to a sample of a new process that has been developed
to reduce the variability of impurities.
Sample 1 Sample 2
n 25 25

237
 Chapter 5 : Hypothesis Testing

s2 1.04 0.51

7. A manufacturer wishes to determine whether there is less variability in the

silver plating done by Company 1 than that done by Company 2. Independent
random samples yield the following results. Do the populations have different
variances?
Sample 1 Sample 2
n 12 12
s2 0.035 mil 0.062 mil

8. A professor has two classes, X and Y. Class X had 13 students and class Y has
25 students. On the same test, although there was no significant difference in
mean grades, class X had a standard deviation of 10 while class Y had a
standard deviation of 13. We can conclude at 1% level of significance, that the
variability of class Y is greater than that of X ?

Answer Exercise 5.8

1. Do not reject the H0.
2. Do not reject the H0.
3. Reject H0.
4. Reject H0.
5. Reject H0.
6. Reject H0.
7. Reject H0.
8. Do not reject H0.

238
 Chapter 5 : Hypothesis Testing

EXERCISE CHAPTER 5
1. Resting pulse rate is an important measure of the fitness of a person’s
cardiovascular system with a lower rate (greater fitness). The mean pulse rate
for all adult males is approximately 72 beats per minute. A random sample of
25 male students currently enrolled in the Faculty of Science was selected and
the mean pulse rate resting pulse rate was found to be 80 beats per minute
with a standard deviation of 20 beats per minute. The experimenter wishes to
test if the students are less fit, on average, than the general population.
(a) What are the null and alternative hypothesis ?
(b) Is there any evidence to support the claim at   0.05 ?
(c) Is there any evidence to support the claim at   0.001 ?

2. The average time it takes for a person to experience pain relief from aspirin is
25 minutes. A new ingredient is added to help speed up relief. Let  denote
the average time to obtain pain relief with the new product. An experiment is
conducted to verify if the new product is better. A random sample of forty
patients in a certain hospital was selected and the mean time for a person
relieved from aspirin was found to be 23 minutes with a standard deviation of
five minutes.
(a) What are the null and alternative hypothesis ?
(b) Is there any evidence to support the claim at α = 0.05 ?
(c) Is there any evidence to support the claim at α = 0.05 ?

3. In order to study the harmful effects of DDT poisoning, the pesticide was fed
to 40 randomly chosen rats out of a group of 80 rats. The other 40 rats were
used as the control group. Table shows the summary of the study about the
amount of tremor detected in the bodies of each rat after the experiment. A
biologist claims that the average tremors of the experiment group (fed with
pesticide) exceed the average tremors of the control group by less than seven
times. Assume that both variances are unknown.

239
 Chapter 5 : Hypothesis Testing

Poisoned Control

x p  7.6 xc  9.483

s p  6.313 sc  1.973

n p  40 nc  40

(a) State the null and alternative hypotheses ?

(b) Is there any evidence to support the claim at α = 0.1 ?

4. Nine birds and ten cats were tested to determine if there is a difference in the
average number of days that the animal can survive without food. The birds
averaged 11 days with a standard deviation of 2 days while the cats averaged
12 days with a standard deviation of 3 days. What can be concluded at 0.01
level of significance ? Assume that the population variances are equal but
unknown.

5. Two different lighting techniques are compared by measuring the intensity of

light at selected locations in areas lighted by the two methods. If 15
measurements in the first area had a standard deviation of 2.7 foot-candles and
21 measurements in the second area had a standard deviation of 4.2 foot-
candles, can it be conclude that the lighting in the second area is less uniform
? Use a 0.01 level of significance.

6. A powder diet is tested on 49 people and a liquid diet is tested on 36 different

people. Of interest is whether the liquid diet yields a higher average weight
loss than the powder diet. The powder diet group had an average weight loss
of 42 pounds with a standard deviation of 12 pounds. The liquid diet group
had an average weight loss of 45 pounds with a standard deviation of 14
pounds. Use a 0.1 level of significance to test the claim.

7. The average number of English courses taken in a two–year time period by

240
 Chapter 5 : Hypothesis Testing

male and female college students is believed to be about the same. An

experiment is conducted and data are collected from 29 males and 16 females.
The males took an average of 3 English courses with a standard deviation of
0.8. The females took an average of 4 English courses with a standard
deviation of 1.0. Are the averages statistically the same? Use a 0.05 level of
significance to test the claim.

8. A study is done to determine if students in the California state university

system take longer to graduate than students enrolled in private universities.
100 students from both the California state university system and private
universities are surveyed. Suppose that from years of research, it is known
that the population standard deviations are 1.5811 years and 1 year,
respectively. The following data are collected. The California state university
system students took on average 4.5 years with a standard deviation of 0.8.
The private university students took on average 4.1 years with a standard
deviation of 0.3. Use a 0.01 level of significance to test the claim.

9. A recent drug survey showed an increase in use of drugs and alcohol among
local high school seniors as compared to the national percent. Suppose that a
survey of 100 local seniors and 100 national seniors is conducted to see if the
percentage of drug and alcohol use is higher locally than nationally. Locally,
65 seniors reported using drugs or alcohol within the past month, while 60
national seniors reported using them. Use a 0.05 level of significance to test
the claim.

10. A student at a four-year college claims that average enrollment at four–year

colleges is higher than at two–year colleges in the United States. Two surveys
are conducted. Of the 35 two–year colleges surveyed, the average enrollment
was 5068 with a standard deviation of 4777. Of the 35 four-year colleges
surveyed, the average enrollment was 5466 with a standard deviation of 8191.

241
 Chapter 5 : Hypothesis Testing

(Source: Microsoft Bookshelf). Use a 0.02 level of significance to test the

claim.

11. We are interested in whether the percents of female suicide victims for ages
15 to 24 are the same for the white and the black races in the United States.
We randomly pick one year, 1992, to compare the races. The number of
suicides estimated in the United States in 1992 for white females is 4930. 580
were aged 15 to 24. The estimate for black females is 330. 40 were aged 15
to 24. We will let female suicide victims be our population. (Source: the
National Center for Health Statistics, U.S. Dept. of Health and Human
Services). Use a 0.01 level of significance to test the claim.

12. At Rachel’s 11th birthday party, 8 girls were timed to see how long (in
seconds) they could hold their breath in a relaxed position. After a two-
minute rest, they timed themselves while jumping. The girls thought that the
jumping would not affect their times, on average. Test their hypothesis by
using a 0.01 level of significance.
Relaxed time (seconds) Jumping time (seconds)
26 21
47 40
30 28
22 21
23 25
45 43
37 35
29 32

13. Elizabeth Mjelde, an art history professor, was interested in whether the value
from the Golden Ratio formula ((larger + smaller dimension)/larger
dimension) was the same in the Whitney Exhibit for works from 1900 – 1919

242
 Chapter 5 : Hypothesis Testing

as for works from 1920 – 1942. 37 early works were sampled. They
averaged 1.74 with a standard deviation of 0.11. 65 of the later works were
sampled. They averaged 1.746 with a standard deviation of 0.1064. Do you
think that there is a significant difference in the Golden Ratio calculation?
(Source: data from Whitney Exhibit on loan to San Jose Museum of Art). Use
a 0.01 level of significance to test the claim.

14. One of the questions in a study of marital satisfaction of dual–career couples

was to rate the statement, “I’m pleased with the way we divide the
responsibilities for childcare.” The ratings went from 1 (strongly agree) to 5
(strongly disagree). Below are ten of the paired responses for husbands and
wives. Conduct a hypothesis test to see if the average difference in the
husband’s versus the wife’s satisfaction level is negative (meaning that, within
the partnership, the husband is happier than the wife). Use a 0.05 level of
significance to test the claim.

Wife’s score 2 2 3 3 4 2 1 1 2 4
Husband’s score 2 2 1 3 2 1 1 1 2 4

15. Ten individuals went on a low–fat diet for 12 weeks to lower their cholesterol.
Evaluate the data below. Do you think that their cholesterol levels were significantly
lowered? Use a 0.1 level of significance to test the claim.

Starting cholesterol level Ending cholesterol level

140 140
220 230
110 120
240 220
200 190
180 150
190 200

243
 Chapter 5 : Hypothesis Testing

360 300
280 300
260 240

16. Eight runners were convinced that the average difference in their individual
times for running one mile versus race walking one mile was at most 2
minutes. Below are their times. Do you agree that the average difference is at
most 2 minutes? Use a 0.02 level of significance to test the claim.

Running time (minutes) Race walking time (minutes)

5.1 7.3
5.6 9.2
6.2 10.4
4.8 6.9
7.1 8.9
4.2 9.5
6.1 9.4
4.4 7.9

17. Marketing companies have collected data implying that teenage girls use more ring
tones on their cellular phones than teenage boys do. In one particular study of 40
randomly chosen teenage girls and boys (20 of each) with cellular phones, the
average number of ring tones for the girls was 3.2 with a standard deviation of 1.5.
The average for the boys was 1.7 with a standard deviation of 0.8. Conduct a
hypothesis test to determine if the averages are approximately the same or if the girls’
average is higher than the boys’ average. Use a 0.01 level of significance to test
the claim.

18. Parents of teenage boys often complain that auto insurance costs more, on
average, for teenage boys than for teenage girls. A group of concerned parents

244
 Chapter 5 : Hypothesis Testing

examines a random sample of insurance bills. The average annual cost for 36
teenage boys was $679. For 23 teenage girls, it was $559. From past years, it
is known that the population standard deviation for each group is $180.
Determine whether or not you believe that the average cost for auto insurance
for teenage boys is greater than that for teenage girls. Use a 0.05 level of
significance to test the claim.

19. A group of transfer bound students wondered if they will spend the same
average amount on texts and supplies each year at their four-year university as
they have at their community college. They conducted a random survey of 54
students at their community college and 66 students at their local four-year
university. The sample means were $947 and $1011, respectively. The
population standard deviations are known to be $254 and $87, respectively.
Conduct a hypothesis test to determine if the averages are statistically the
same by using a 0.1 level of significance.

20. Some manufacturers claim that non-hybrid sedan cars have a lower average
miles per gallon (mpg) than hybrid ones. Suppose that consumers test 21
hybrid sedans and get an average 31 mpg with a standard deviation of 7 mpg.
Thirty-one non-hybrid sedans average 22 mpg with a standard deviation of 4
mpg. Suppose that the population standard deviations are known to be 6 and
3, respectively. Conduct a hypothesis test to the manufacturers claim by using
a 0.01 level of significance.

ANSWER EXERCISE CHAPTER 5

1. (a) H 0 :   72 , H1 :   72 (Claim) (b) Reject the H0.
(c) Do not reject the H0.
2. (a) H 0 :   25 , H1 :   25 (Claim) (b) Reject the H0.
(c) Do not reject the H0.
3. (a) H 0 : P  C  7 , H1 : P  C  7 (Claim)

245
 Chapter 5 : Hypothesis Testing

(b) Reject the H0.

4. Do not reject the H0. 5. Do not reject the H0.

SUMMARY CHAPTER 5

Type I and Type II errors can be defined in terms of hypothesis testing.

 A Type I error (α) is the probability of rejecting a true null hypothesis.
 A Type II error (β) is the probability of failing to reject a false null hypothesis.

Hypothesis testing steps

Here are the steps to performing hypothesis testing
Step 1
Write the original claim and identify whether it is the null hypothesis or the
alternative hypothesis.
Step 2
Use the alternative hypothesis to identify the type of test.
Write down all information from the problem and with specific case.
State the distribution should use. Find the critical value using the tables and state the
decision rule.

Step 3
Compute the test statistic.

Step 4
Make a decision to reject or fail to reject the null hypothesis. A picture showing the
critical value and test statistic may be useful.

Step 5
Write the conclusion.

246
 Chapter 5 : Hypothesis Testing

Case Variances Samples size Statistical Test

A Known n1 , n2  30 ( X 1  X 2 )  ( 1   2 )
Z Test 
 12  22

n1 n2

B Known n1 , n2  30 ( X 1  X 2 )  ( 1   2 )
Z Test 
 12  22

n1 n2

C Unknown n1 , n2  30 ( X 1  X 2 )  ( 1   2 )
Z Test 
s12 s2
 2
n1 n2

D Unknown (Equal) n1 , n2  30 ( X 1  X 2 )  ( 1   2 )
TTest 
1 1
Sp  
n1 n 2

v  n1  n2  2
E Unknown (Not equal) n1  n 2  30 ( X 1  X 2 )  ( 1   2 )
TTest 

1 2
n
s1  s 22 

v  2(n  1)
F Unknown (Not equal) n1 , n2  30 ( X 1  X 2 )  ( 1  2 )
TTest 
s12 s2
 2
n1 n2
2
 s12 s22 
  
v   12
n n2 
2
 s12   s22 
   
 n1    n2 
n1  1 n2  1

247
 Chapter 5 : Hypothesis Testing

CORRECTION PAGE CHAPTER 5

248