Republic of the Philippines

Cagayan State University

Carig Campus
COLLEGE OF ENGINEERING
CSU Vision

Transforming
lives by
Educating for FLIPPED NOTES NUMBER 9
the BEST.

CSU Mission

CSU is committed
to transform the In partial fulfilment for the requirements of the course
lives of people and ENGINEERING DATA ANALYSIS
communities
through high
quality instruction
and innovative
research,
development, By:
production and
SUBONG, JOEMAR D.
extension.
BACANI, VALERIE ELAINE M.
DOCA, AL JOHNKENETH A.
CSU – IGA
Competence
TANNAGAN, NOREEN G.
Social Responsibility

Unifying Presence

COE – IGA
Innovative Thinking
January 06, 2020
Synthesis
Personal
Responsibility
Empathy
Research Skill
Entrepreneurial Skill
 Applications of Testing Hypothesis

Learning Outcomes

By the end of this chapter, the following targets should have been accomplished

 apply the general steps of hypotheses testing concerning different parameters of a

population mean when the sample size is large and small.
 interpret the result of a test of hypotheses in the context of the given situation.

I. Introduction

This chapter presents applications of the hypothesis-testing logic discussed in

Chapter 8. Among the population parameters to be considered are

Parameter Description
  Mean; average
 Difference in means or averages; mean difference;
(1  2) comparison of means or averages

p  Proportion; percentage; fraction; rate

 Difference in proportion, percentage, fraction, or rates;
(p1  p2) comparison of proportions, percentages, fractions, or rates

2  Variance; variation; precision

σ 12  Ratio of variances; difference in variation; comparison of
σ 22 variances

In all hypothesis testing of parameters, the same general processes are employed:
stating the null and alternative hypotheses, determining and computing test statistic,
establishing rejection region, making decisions and interpreting the decisions.
 II. Hypothesis Test about a Population Mean

By the end of this chapter, the following targets should have been accomplished

 apply the general steps of hypothesis testing to identify significant difference

between a population mean and a single statistic value when the sample size is
large or small.
 interpret the result of a test of hypotheses in the context of the given situation.

a. Tests of population means using large samples

In this section we describe and demonstrate the procedure for conducting a test of
hypotheses about the mean of a population in the case that the sample size n is at least 30.
The Central Limit Theorem states that X́ is approximately normally distributed, and has
mean μ x́ =μ and standard deviation σ x́ =σ / √n, where μ and σ are the mean and the
standard deviation of the population. This implies that the statistic
x́−μ
σ /√n
has a normal standard distribution.
Large-sample test of hypothesis about a population mean
ONE -TAILED TEST TWO -TAILED TEST
H0:  = 0 H0:  = 0
Ha:  > 0 (or Ha:  < 0) Ha:   0

Test statistic if σ is known Test statistic if σ is unknown

x́ −μ 0 x́ −μ 0
Z= Z=
σ /√n s / √n

Rejection region: Rejection region:

z > z (or z < - z) z < -z/2 (or z > z/2)

where z is the z-value such that P(z > z) = ; and z/2 is the z-value such that P(z > z/2) = /2.
[Note: 0 is our symbol for the particular numerical value specified for  in the null hypothesis.]

Assumption: The sample size must be sufficiently large (say, n  30) so that the sampling
distribution of x̄ is approximately normal and that s provides a good approximately to .
Example 1
It is hoped that a newly developed pain reliever will more quickly produce
perceptible reduction in pain to patients after minor surgeries than a standard pain reliever.
The standard pain reliever is known to bring relief in an average of 3.5 minutes with
standard deviation 2.1 minutes. To test whether the new pain reliever works more quickly
than the standard one, 50 patients with minor surgeries were given the new pain reliever
and their times to relief were recorded. The experiment yielded sample mean x́=3.1
minutes and sample standard deviation s=1.5 minutes. Is there sufficient evidence in the
sample to indicate, at the 5% level of significance, that the newly developed pain reliever
does deliver perceptible relief more quickly?

Solution
Step 1. The natural assumption is that the new drug is no better than the old one, but must
be proved to be better. Thus if μ denotes the average time until all patients who are given
the new drug experience pain relief, the hypothesis test is
H 0 : μ=3. 5
H a : μ<3 . 5
Step 2. The sample is large, but the population standard deviation is unknown (the 2.1
minutes pertains to the old drug, not the new one). Thus the test statistic is
x−μ 0
Z=
s
√n
and has the standard normal distribution.
Step 3. Solving for the test statistic
x−μ 0 3. 1−3 . 5
Z= = =−1 .886
s 1 .5
√n √ 50
Step 4. Since the symbol in H a is “<” this is a left-tailed test, so there is a single critical
value, −z α =−z 0.05=−1.645. The rejection region is (−∞,−1.645]
Step 5. As shown in Figure 1 the test statistic falls in the rejection region. The decision is
to reject H0.
 Figure 1. Region and Test Statistic for Example 1
In the context of the problem our conclusion is:
The data provide sufficient evidence, at the 5% level of significance, to conclude
that the average time until patients experience perceptible relief from pain using the new
pain reliever is smaller than the average time for the standard pain reliever.

Example 2 A cosmetics company fills its best-selling 8 ounce jars of facial cream by an
automatic dispensing machine. The machine is set to dispense a mean of 8.1 ounces per
jar. Uncontrollable factors in the process can shift the mean away from 8.1 and cause
either under fill or overfill, both of which are undesirable. In such a case the dispensing
machine is stopped and recalibrated. Regardless of the mean amount dispensed, the
standard deviation of the amount dispensed always has value 0.22 ounce. A quality control
engineer routinely selects 30 jars from the assembly line to check the amounts filled. On
one occasion, the sample mean is x́=¿8.2 ounces and the sample standard deviation is
s=0.25 ounce. Determine if there is sufficient evidence in the sample to indicate, at the 1%
level of significance, that the machine should be recalibrated.
Solution
Step 1. The natural assumption is that the machine is working properly. Thus if μ denotes
the mean amount of facial cream being dispensed, the hypothesis test is
H 0 : μ=8. 1
H a : μ≠8 .1
Step 2. The sample is large and the population standard deviation is known. Thus the test
statistic is
x−μ 0
Z=
σ
√n
and has the standard normal distribution.
Step 3. Solving for the test statistic
x−μ 0 8 .2−8 .1
Z= = =2 . 490
σ 0 . 22
√n √ 30
Step 4. Since the symbol in H a is “≠” this is a two-tailed test, so there are two critical
values, −z α / 2=± z0.005 =± 2.576. The rejection region is (−∞,−2.576]∪[2.576,∞).
Step 5. As shown in Figure 2 the test statistic does not fall in the rejection region. The
decision is to accept H0.

Figure 2. Region and Test Statistic for Example 2

In the context of the problem our conclusion is:
The data do not provide sufficient evidence, at the 1% level of significance, to
conclude that the average amount of product dispensed is different from 8.1 ounce. We
conclude that the machine does not need to be recalibrated.

The p-value Approach to Hypothesis Testing

This approach just uses the corresponding probability or area under the normal
distribution curve instead of the z values. Considering Example 2, the −z α / 2=± z0.005 =
± 2.576 and the rejection area is ¿ ∪ [ 2.576 , ∞ ) which corresponds to a total area of α
=0.01. The z statistic may or may not fall on the rejection area will have an analogous area
of ¿ ∪ [ z , ∞ ) which corresponds to an area we shall call p. In this case, if z statistic falls on
the rejection area, it can be concluded that α > p. Hence, we reject H0. In the same way, if z
statistic is not on the rejection area, it can be concluded that α < p. Hence we accept H0. In
example 2, the value of the test statistic was z=2.490, with a corresponding area of p=¿
0.00640×2=0.0128, since the test was two-tailed. In Example 2 the test was performed at
the 1% level of significance or α=0.01. Since p=0.0128>0.01 (or 1.28% is greater than
1%), the decision is to accept null hypothesis. This decision is also the same using the
method in example 2. This p-value approach is convenient when using statistical software
like SPSS, Microsoft Excel and the like.
 Tests of population means using small samples
When sample sizes are small, as is often the case in practice, the Central Limit
Theorem does not apply. One must then impose stricter assumptions on the population to
give statistical validity to the test procedure. One common assumption is that the
population from which the sample is taken has a normal probability distribution to begin
with. If σ is unknown and is approximated by the sample standard deviation s, then the
resulting test statistic t follows Student’s t-distribution with n−1 degrees of freedom.
Small-sample test of hypothesis about a population mean
ONE-TAILED TEST TWO-TAILED TEST
H0:  = 0 H0:  = 0
Ha:  > 0 (or Ha:  < 0) Ha:   0
Test statistic:
x  0
t
s/ n
Rejection region: Rejection region:
t > t (or t < - t) t < -t/2 (or t > t/2)

where the distribution of t is based on (n – 1) degrees of freedom; t is the t-value such that P(t
> t) = ; and t/2 is the t-value such that P(t > t/2) = /2.

Assumption: The relative frequency distribution of the population from Which the sample was
selected is approximately normal.

Example 3
The price of a popular tennis racket at a national chain store is $179. Portia bought five of
the same racket at an online auction site for the following prices:
155 179 175 175 161
Assuming that the auction prices of rackets are normally distributed, determine whether
there is sufficient evidence in the sample, at the 5% level of significance, to conclude that
the average price of the racket is less than $179 if purchased at an online auction.

Solution
Step 1. The assertion for which evidence must be provided is that the average online price
μ is less than the average price in retail stores, so the hypothesis test is
H 0 : μ=179
H a : μ<179
Step 2. The sample is small and the population standard deviation is unknown. Thus the
test statistic is
x −μ 0
t=
σ
√n
and has the Student t-distribution with n−1=5−1=4 degrees of freedom.
Step 3. From the data we compute x́=169 and s=10.39. Inserting these values into the
formula for the test statistic gives
x −μ 0 169−179
t= = =−2. 152
σ 10 . 39
√n √5
Step 4. Since the symbol in Ha is “<” this is a left-tailed test, so there is a single critical
value, −t α =−t 0.05[df=4]. Reading from the row labeled df=4 in the critical t table its value is
−2.132. The rejection region is (−∞,−2.132].
Step 5. As shown in Figure 3, the test statistic falls in the rejection region. The decision is
to reject H0.

Figure 3. Region and Test Statistic for Example 3

In the context of the problem our conclusion is:
The data provide sufficient evidence, at the 5%level of significance, to conclude that the
average price of such rackets purchased at online auctions is less than $179.
Example 4
A small component in an electronic device has two small holes where another tiny
part is fitted. In the manufacturing process the average distance between the two holes
must be tightly controlled at 0.02 mm, else many units would be defective and wasted.
Many times throughout the day quality control engineers take a small sample of the
components from the production line, measure the distance between the two holes, and
make adjustments if needed. Suppose at one time four units are taken and the distances are
measured as
0.021 0.019 0.023 0.020
Determine, at the 1% level of significance, if there is sufficient evidence in the sample to
conclude that an adjustment is needed. Assume the distances of interest are normally
distributed.
Solution
Step 1. The natural assumption is that the machine is working properly. Thus if μ denotes
the mean amount of facial cream being dispensed, the hypothesis test is
H 0 : μ=0. 02
H a : μ≠0 .02
where μ denotes the mean distance between the holes.
Step 2. The sample is small and the population standard deviation is unknown. Thus the
test statistic is
x −μ 0
t=
s
√n
and has the Student t-distribution with n−1=4−1=3 degrees of freedom.
Step 3. From the data we compute x́=0.02075 and s=0.00171. Inserting these values into
the formula for the test statistic gives
x −μ 0 0 . 02075−0 . 02
t= = =0 . 877
s 0 .00171
√n √4
Step 4. Since the symbol in H a is “≠” this is a two-tailed test, so there are two critical
values, −t α / 2=± t 0.005 [df=3]= ±5.841. The rejection region is (−∞,−5.841]∪[5.841,∞).
Step 5. As shown in Figure 4 the test statistic does not fall in the rejection region. The
decision is to accept H0.
 Figure 4. Region and Test Statistic for Example 4
In the context of the problem our conclusion is:
The data do not provide sufficient evidence, at the 1% level of significance, to
conclude that the mean distance between the holes in the component differs from 0.02
mm.
 III. Hypothesis tests of population proportions

By the end of this chapter, the following targets should have been accomplished

 apply the general steps of hypothesis testing to identify significant difference

between a population proportion and a single statistic proportion when the sample
size is large or small.
 interpret the result of a test of hypotheses in the context of the given situation.

Large-sample test of hypothesis about a population proportion

The procedure used to test a hypothesis about a population proportion, p, is based

on a large sample from the target population. (Recall that p represents the probability of
success in a Bernoulli process.)

In order that the procedure to be valid, the sample size must be sufficiently large to
guarantee approximate normality of the sampling distribution of the sample proportion, p.
A general rule of thumb for determining whether n is "sufficiently large" is that the
interval ^p ± 2 √ ^p q^ /n does not include 0 or 1.

Large-sample test of hypothesis about a population proportion

ONE -TAILED TEST TWO -TAILED TEST
H0: p = p0 H0: p = p0
Ha: p > p0 (or Ha: p < p0) Ha: p  p0
Test statistic:
pˆ  p0
z
p0 q0 / n

Rejection region: Rejection region:

z > z (or z < - z) z < -z/2 (or z > z/2)
where q0 = 1 – p0 where q0 = 1 – p0

Assumption: The interval ^p ± 2 √ ^p q^ /n does not contain 0 and 1.

Example 5

A soft drink maker claims that a majority of adults prefer its leading beverage over
that of its main competitor’s. To test this claim 500 randomly selected people were given
the two beverages in random order to taste. Among them, 270 preferred the soft drink
maker’s brand, 211 preferred the competitor’s brand, and 19 could not make up their
minds. Determine whether there is sufficient evidence, at the 5% level of significance, to
support the soft drink maker’s claim against the default that the population is evenly split
in its preference.

Solution:

We will use the critical value approach to perform the test. We must check that the sample
is sufficiently large to validly perform the test. Since ^p=270/500=0.54

[ √p^ −3
^p q^
n
, p^ +3
n√ ]
p^ q^

[
= 0.54−3
√ 0. 54×0.46
500
,0.54+3
0.54×0.46
500 √ ]
=[ 0.48,0.60 ] ⊂ [ 0,1 ]

so the sample is sufficiently large.

Step 1. The relevant test is

H 0 : p=0. 50
H a : p>0 . 50
where p denotes the proportion of all adults who prefer the company’s beverage over that
of its competitor’s beverage.

Step 2. The test statistic is

^p −p 0
z=
p0 q 0
√ n
and has the standard normal distribution.
Step 3. The value of the test statistic is
^p −p 0 0 .54−0. 50
z= = =1 .789
p0 q 0 0 .50×0 . 50
√ √
n 500
Step 4. Since the symbol in H a is “¿” this is a right-tailed test, so there is a critical value,
z α = z 0.005[df=3]= 1.645. The rejection region is [1.645,∞).

Step 5. As shown in Figure 5 the test statistic falls in the rejection region. The decision is
to reject H0.

Figure 5. Region and Test Statistic for Example 5

In the context of the problem our conclusion is:
The data provide sufficient evidence, at the 5% level of significance, to conclude
that a majority of adults prefer the company’s beverage to that of their competitor’s.
 IV. Hypothesis tests about the difference between two population means

By the end of this chapter, the following targets should have been accomplished

 apply the general steps of hypothesis testing to identify significant difference

between two population means when the sample sizes are large or small.
 interpret the result of a test of hypotheses in the context of the given situation.

a. Comparison of Two Population Means: Large, Independent Samples

Suppose we wish to compare the means of two distinct populations. From the two
populations, we take large two independent sample as permitted by the Central Limit
Theorem. Samples from two distinct populations are said to be independent if each one is
drawn without reference to the other, and has no connection with the other. Our goal is to
use the information in the samples to estimate the difference μ1−μ2 in the means of the two
populations and to make statistically valid inferences about it.

Large-sample test of hypothesis about (1 - 2)

ONE -TAILED TEST TWO -TAILED TEST
H0: (1 - 2) = D0 H0: (1 - 2) = D0
Ha: (1 - 2) > D0 (or Ha: (1 - 2)< D0) Ha: (1 - 2)  D0

Test statistic:

( x 1  x2 )  D0 ( x1  x 2 )  D0
z 
 ( x1  x2 ) s1 2 s2 2

n1 n2

Rejection region: Rejection region:

z > z (or z < - z) z < -z/2 or z > z/2

[Note: In many practical applications, we wish to hypothesize that there is no difference

between the population means; in such cases, D0 = 0]

Assumptions:
1. The sample sizes n1 and n2 are sufficiently large (n1  30 and n2  30).
2. The samples are selected randomly and independent from the target populations.

Example 6
To compare customer satisfaction levels of two competing cable television
companies, 174 customers of Company 1 and 355 customers of Company 2 were
randomly selected and were asked to rate their cable companies on a five-point scale, with
1 being least satisfied and 5 most satisfied. The survey results are summarized in the
following table:
Company 1 Company 2
n1 = 174 n2 = 355
 x́ 1= 3.51 x́ 2= 3.24
s1 = 0.51 s2 = 0.52
Test at the 1% level of significance whether the data provide sufficient evidence to
conclude that Company 1 has a higher mean satisfaction rating than does Company 2. Use
the critical value approach.
Solution:

Step 1. The relevant test is

H 0 : μ1 −μ2 =0
H a : μ1 −μ2 > 0
Step 2. Since the samples are independent and both are large the test statistic is

( x̄ 1− x̄ 2 )−D0
z=
s12 s22
√ +
n1 n2

Step 3. The value of the test statistic is

( x̄ 1− x̄ 2 )−D0 ( 3 . 51−3 . 24 )−0

z= = =5 .684
s12 s22 0. 512 0 .522
√ +
n1 n2 √ 174
+
355

Step 4. Since the symbol in H a is “¿” this is a right-tailed test, so there is a critical value,
z α = z 0.01[df=3]= 2.326 The rejection region is [2.326,∞).

Step 5. As shown in Figure 6 the test statistic falls in the rejection region. The decision is
to reject H0.

Figure 6. Region and Test Statistic for Example 6

In the context of the problem our conclusion is:
The data provide sufficient evidence, at the 1% level of significance, to conclude
that the mean customer satisfaction for Company 1 is higher than that for Company 2.
 b. Comparison of Two Population Means - Small, Independent Samples
When one or the other of the sample sizes is small, as is often the case in practice,
the Central Limit Theorem does not apply. We must then impose conditions on the
population to give statistical validity to the test procedure. We will assume that both
populations from which the samples are taken have a normal probability distribution and
that their standard deviations are equal.

Small-sample test of hypothesis about (1 - 2)

ONE -TAILED TEST TWO -TAILED TEST
H0: (1 - 2) = D0 H0: (1 - 2) = D0
Ha: (1 - 2) > D0 (or Ha: (1 - 2)< D0) Ha: (1 - 2)  D0

Test statistic:

( x 1  x2 )  D0
t
 1 1 
s 2p   
 n1 n2 
Rejection region: Rejection region:
t > t (or t < - t) t < -t/2 or t > t/2

Where

(n1  1)s12  (n2  1)s22

s2p 
n1  n2  2

and the distribution of t is based on (n1 + n2 - 2) degrees of freedom.

Assumptions:
1. The population from which the samples are selected both have approximately normal
relative frequency distributions.
2. The variances of the two populations are equal.
3. The random samples are selected in an independent manner from the two populations.

Example 7
A software company markets a new computer game with two experimental
packaging designs. Design 1 is sent to 11 stores; their average sales the first month is 52
units with sample standard deviation 12 units. Design 2 is sent to 6 stores; their average
sales the first month is 46 units with sample standard deviation 10 units. Test at the 1%
level of significance whether the data provide sufficient evidence to conclude that the
mean sales per month of the two designs are different. Use the critical value approach.
Solution:

Step 1. The relevant test is

H 0 : μ1 −μ2 =0
H a : μ1 −μ2 > 0
Step 2. Since the samples are independent and at least one is less than 30 the test statistic
is

( x̄1 − x̄ 2 ) −D0
t=
1 1
√ s 2p
( +
n1 n 2 )
which has Student’s t-distribution with df=11+6−2=15 degrees of freedom.

( n1−1 ) s 21 + ( n2 −1 ) s 22 ( 10 ) (12 )2 + ( 5 ) (10 )2

s 2p = = =129 .3
n1 + n2 −2 15

Step 3. The value of the test statistic is

( x̄1 − x̄ 2 ) −D0 ( 52−46 )−0

t= = =1 . 040
1 1 1 1
√ s 2p
( +
n1 n 2 ) √ 129 .3 +
11 6( )
Step 4. Since the symbol in H a is “≠” this is a two-tailed test, so there are two critical
values, ± t α /2=± t 0.005 [df=15]= 2.947 The rejection region is (−∞,−2.947]∪[2.947,∞).

Step 5. As shown in Figure 7 the test statistic does not fall in the rejection region. The
decision is to accept H0.

Figure 7. Region and Test Statistic for Example 7

 In the context of the problem our conclusion is:
The data do not provide sufficient evidence, at the 1% level of significance, to
conclude that the mean sales per month of the two designs are different.
V. Hypothesis tests about the difference between two proportions

By the end of this chapter, the following targets should have been accomplished

 apply the general steps of hypothesis testing to identify significant difference

between two population proportions when the sample size is large or small.
 interpret the result of a test of hypotheses in the context of the given situation.

Suppose we wish to compare the proportions of two populations that have a

specific characteristic, we take large two independent sample as permitted by the Central
Limit Theorem. Samples from two distinct populations are said to be independent if each
one is drawn without reference to the other, and has no connection with the other. Our
goal is to use the information in the samples to estimate the difference in proportions of
the two populations and to make statistically valid inferences about it.
Large-sample test of hypothesis about (p1 - p2)
ONE -TAILED TEST TWO -TAILED TEST
H0: (p1 - p2) = D0 H0: (p1 - p2) = D0
Ha: (p1 - p2) > D0 or (Ha: (p1 - p2) < D0) Ha: (p1 - p2)  D0

Test statistic:

( pˆ 1  pˆ 2 )  D0
z
 ( pˆ 1  pˆ2 )
 Rejection region: Rejection region:
z > z (or z < - z) z < -z/2 or z > z/2

p1 q1 p2q2
 ( pˆ1  pˆ2 )  
n1 n2
where

when D0  0, calculate  ( pˆ 1  pˆ2 ) using pˆ 1 and pˆ2 :

pˆ 1 qˆ1 pˆ 2qˆ2
 ( pˆ1  pˆ2 )  
n1 n2

where qˆ1  1  pˆ 1 and qˆ2  1  pˆ 2 .

For the special case where D0 = 0, calculate

 1 1 
 ( pˆ1  pˆ2 )  pq
ˆ ˆ  
 n1 n2 
when the total number of successes in the combined samples is (x1 + x2) and
x 1  x2
pˆ 1  pˆ 2  pˆ  .
n1  n2

Assumption: The intervals

pˆ 1  2 pˆ 1 qˆ1 / n1 and pˆ 2  2 pˆ 2 qˆ2 / n2
do not contain 0 and 1.
Example 8
The department of code enforcement of a county government issues permits to
general contractors to work on residential projects. For each permit issued, the department
inspects the result of the project and gives a “pass” or “fail” rating. A failed project must
be re-inspected until it receives a pass rating. The department had been frustrated by the
high cost of re-inspection and decided to publish the inspection records of all contractors
on the web. It was hoped that public access to the records would lower the re-inspection
rate. A year after the web access was made public, two samples of records were randomly
selected. One sample was selected from the pool of records before the web publication and
one after. The proportion of projects that passed on the first inspection was noted for each
sample. The results are summarized below.
No public web access n1=500 ^p1=0.67
Public web access n2=100 ^p2=0.80

Test whether there is sufficient evidence to conclude that public web access to the
inspection records has increased the proportion of projects that passed on the first
inspection by more than 5 percentage points. Use the critical value approach at the 10%
level of significance.

Solution:
We will use the critical value approach to perform the test. We must check that the
samples are sufficiently large to validly perform the test.
For sample 1,

[ √
p^ −3
^p q^
n
, p^ +3
p^ q^
n √ ]
[
= 0. 67−3
√ 0. 67×0. 33
500
,0 .67+3
0. 67×0.33
500 √ ]
=[ 0.61,0.73 ] ⊂ [ 0,1 ]

so the sample is sufficiently large.

For sample 2,

[ √
p^ −3
^p q^
n √ ]
, p^ +3
p^ q^
n

[ √
= 0.8−3
0.8×0 .2
100
,0.8+3
√ ]
0.8×0.2
100
=[ 0.68,0.92 ] ⊂ [ 0,1 ]

so the sample is also sufficiently large

Step 1. The relevant test is

H 0 : p1 −p 2 =−0 .050
H a : p1 − p2 <0. 050
Step 2. The test statistic is

( ^p1 − p^ 2 ) −D0
z=
^p1 q^ 1 ^p2 q^ 2
√ n1
+
n2

Step 3. The value of the test statistic is

( ^p1 − p^ 2 ) −D0 ( 0. 67−0 . 8 )−(−0 . 05 )

z= = =−1. 770
^p1 q^ 1 ^p2 q^ 2 0 .67×0 . 33 0 . 80×0 . 20
√ n1
+
n2 √ 500
+
100

Step 4. Since the symbol in H a is “¿” this is a left-tailed test, so there is a critical value, z α
= z 0.10= -1.282. The rejection region is (−∞,−1.282].
Step 5. As shown in Figure 8 the test statistic falls in the rejection region. The decision is
to reject H0.

Figure 8. Region and Test Statistic for Example 8

In the context of the problem our conclusion is:
The data provide sufficient evidence, at the 10% level of significance, to conclude
that the rate of passing on the first inspection has increased by more than 5 percentage
points since records were publicly posted on the web.

VI. Hypothesis test about a population variance

By the end of this chapter, the following targets should have been accomplished

 apply the general steps of hypothesis testing to identify significant difference

between a population variance an a single statistic value
 interpret the result of a test of hypotheses in the context of the given situation.

A test of a single variance assumes that the underlying distribution is normal. The
null and alternative hypotheses are stated in terms of the population variance (or
population standard deviation).

Test of hypothesis about a population variance 2

ONE -TAILED TEST TWO -TAILED TEST
H0: 2 = 02 H0: 2 = 02
Ha: 2 > 02 or (Ha: 2 < 02) Ha: 2  02
 Test statistic:

(n  1)s2
2 
 02
One-tailed Two-tailed
Rejection region: Rejection region:
2 > 2 (or 2 < 21-) 2 < 21-/2 or 2 > 2/2
2 2
where χ ∞ and χ 1−∞ are values of 2 that locate an area of  to the right and  to the
left, respectively, of a chi-square distribution based on (n -1) degrees of freedom.
2
[Note: σ 0 is our symbol for the particular numerical value specified for 2 in the null
hypothesis.]

Assumption: The population from which the random sample is selected has an approximate
normal distribution.

Example 9
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 finds 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.

Solution:
Since the claim is that a single line causes less variation, this is a test of a single
variance. The parameter is the population variance, σ2, or the population standard deviation, σ.
Step 1. The relevant test is
2 2
H 0 :σ =7 . 2
2 2
H a : σ <7 . 2
The word "less" tells you this is a left-tailed test.
Step 2. The test statistic is
( n−1 ) s 2
2
χ =
σ2
df=n– 1 = 25 – 1 = 24
Step 3. The value of the test statistic is

2 ( n−1 ) s 2 ( 25−1 ) 3 .5 2
χ = = =5 .67
σ2 7 . 22

Step 4. Since the symbol in H a is “¿” this is a left-tailed test, so there is a critical value,
χ 2α = χ 20.05= 13.85. The rejection region is (0,13.85].

Step 5. As shown in Figure 9 the test statistic falls in the rejection region. The decision is
to reject H0.

α =0.5

1−α=0.95

χ 20.05= 13.85

Figure 9. Region and Test Statistic for Example 9

In the context of the problem our conclusion is:
At a 5% level of significance, from the data, there is sufficient evidence to
conclude that a single line causes a lower variation among the waiting times or with a
single line, the customer waiting times vary less than 7.2 minutes.
VII. Hypothesis test about the ratio of two population variances

By the end of this chapter, the following targets should have been accomplished

 apply the general steps of hypothesis testing to identify significant difference

between two variances
 interpret the result of a test of hypotheses in the context of the given situation.

In some practical situations the difference between the population standard

deviations σ1 and σ2 is also of interest. Standard deviation measures the variability of a
random variable. For example, if the random variable measures the size of a machined part
in a manufacturing process, the size of standard deviation is one indicator of product
quality. A smaller standard deviation among items produced in the manufacturing process
is desirable since it indicates consistency in product quality.

Test of hypothesis for the ratio of two population variances,  1 /  2

2 2

ONE -TAILED TEST TWO -TAILED TEST

2 2 2 2
H0: σ 1 /σ 2 = 1 H0: σ 1 /σ 2 = 1 (i. e . σ 21 =σ 22 )
(i. e . σ 21 =σ 22 )
2 2
Ha: σ 1 /σ 2 ≠ 1 (i. e . σ 21 ≠σ 22 )
2 2
H : σ 1 /σ 2 > 1
a

(i. e . σ 21 >σ 22 ) or
 2 2
[Ha: σ 1 /σ 2 < 1
(i. e . σ 21 < σ 22 ) ]

Test statistic: Test statistic:

2 2
s s Larger sample variance
F 1
2
or F  2
2
F
s
2 s
1
Smaller sample variance i.e.
 s12
 2 when s1  s2
2 2

s
F   22
 s2 when s 2  s2
 s12 2 1

Rejection region: Rejection region:

F > F F > F/2

where F, and F/2 are values that locate an area  and /2, respectively, in the upper tail of
the F-distribution with 1 = numerator degrees of freedom (i.e., the df for the sample
variance in the numerator) and 2 = denominator degrees of freedom (i.e., the df for the
sample variance in the denominator).

Assumptions: 1. Both of the populations from which the samples are selected have relative
frequency distributions that are approximately normal.
2. The random samples are selected in an independent manner from the two
populations.

Example 10
One of the quality measures of blood glucose meter strips is the consistency of the
test results on the same sample of blood. The consistency is measured by the variance of
the readings in repeated testing. Suppose two types of strips, A and B, are compared for
their respective consistencies. We arbitrarily label the population of Type A strips
Population 1 and the population of Type B strips Population 2. Suppose 15 Type A strips
were tested with blood drops from a well-shaken vial and 20 Type B strips were tested
with the blood from the same vial. The results are summarized in the table below. Assume
the glucose readings using Type A strips follow a normal distribution with variance σ 12and
those using Type B strips follow a normal distribution with variance with σ 12. Test, at the
10% level of significance, whether the data provide sufficient evidence to conclude that
the consistencies of the two types of strips are different.
Strip Type Sample Size Sample Variance
A n1=16 s21=2.09
B n2=21 s22=1.10
Solution:
Since the claim is that a single line causes less variation, this is a test of a single
variance. The parameter is the population variance, σ2, or the population standard deviation, σ.
Step 1. The relevant test is
2 2
H 0 : σ 1 =σ 2
2 2
H a : σ 1≠σ 2
Step 2. The test statistic is
2
σ1
F=
σ 22

The distribution is the F-distribution with degrees of freedom df1=16−1=15 and

df2=21−1=20
Step 3. The value of the test statistic is

σ 21 2. 09
F= = =1 . 90
σ 22 1. 10

Step 4. The test is two-tailed. The left or lower critical value is F1−α=F0.95=0.43. The right
or upper critical value is Fα/2=F0.05=2.20. Thus the rejection region is [0,−0.43]∪[2.20,∞).

Step 5. As shown in Figure 10 the test statistic 1.90 does not lie in the rejection region.
The decision is to accept H0.

Figure 10. Region and Test Statistic for Example 10

In the context of the problem our conclusion is:
 The data do not provide sufficient evidence, at the 10% level of significance, to
conclude that there is a difference in the consistency, as measured by the variance, of the
two types of test strips.

References:
Bluman, A. (2013). Elementary Statistics A Step by Step Approach. Macmillan/McGraw-
Hill.

Holmes, A., Illowsky, B., & Dean, S. (2018). Introductory business statistics. Houston,
TX: OpenStax College, Rice University.

Myers, S. L., Ye, K., & Walpole, R. E. (2007). Probability & statistics for engineers &
scientists. 8th Ed. Upper Saddle River, NJ: Pearson Prentice Hall.

Shafer, D. S. & Zhang, Z. (2013). Introductory Statistics. Flat World Knowledge Inc.