Inference regarding the population variances:

An important area of statistics is concerned with making inferences about the

population variance

.

Knowledge of population variability is an important element of statistical
analysis.

For example,

a) A population training method.
It provides low dispersion in trainee productivity levels.
Trainers job is less demanding than a high variability procedure
world.

b) A bank policy favor a single waiting line that feeds into several letter.
may remain same whether more than one line formed.

may be lower in single line.

is higher in more than one line.

c) For a car rental company.
Tires with low variability is preferred compared with durable tire
with high variability replacement.

Like the population mean ,

is ordinarily unknown, and its value must be

estimated using simple data.

Sample Variance:

A random sample of n observations drawn from a population with unknown
mean and unknown variance

. Denote the sample

, -------------------

The population is the expectation

= E [(x-)^2]

Whish suggests that we consider the mean of (

- )^2 over n observations. Since

is unknown, the sample mean is used to calculate a sample variance.



The quantity-

-


Is called a sample variance, and its square root, s, is called the sample standard
deviation.

Given a specific random sample, we could compute the sample variance, and the
sample variance would be different for each random sample because of
differences in sample observations.

n- is known as degrees of freedom.
We will show you that, E(

) =

.

is the consistent and unbiased estimator of .

is the consistent and unbiased estimator of

.

Thus,

serves as the basis either for constructing confidence interval estimates

or for testing hypothesis about

.

The conclusion that the expected value of the sample variance is the population
variance is quite general. But, for statistical inference we would like to know
more about the sampling distribution.

If we can assume that the underlying population distribution is normal, then it
can be shown that the sample variance and the population variance are related
through a probability distribution known as the chi-square distribution.

Multiplying

by n-1 and dividing by

converts it into the chi square random

variable:

,

Which has the chi-square distribution. Thus it is possible to find probabilities for
the variable:

]

When the number of degrees of freedom is n-1.


The Chi-Square Distribution:

For a large sample size, the sampling distribution of chi-square can be closely
approximated by a continuous curve known as the chi-square distribution.

The probability function of a chi-square distribution is given by:

F(

Where e = 2.71828, v = d.f c = a constant expending only on v.



IMPORTANT PROPERTIES OF CHI-SQUARE DISTRIBUTION:

1. Chi-square distribution is a continuous probability distribution which has
the value zero at its lower limit at extends to infinity in the positive
direction.

Negative value of chi-square is not possible since the differences between
the observed and expected frequencies are always squared.

2. The equal shape of the distribution depends upon the number of degrees
of freedom v. For different values of v, we shall have different shapes of
the distribution.

In general, when v is small, the shape of the curve is skewed to the right
and as v gets larger, the distribution becomes more and more
symmetrical and can be approximated by the normal distribution.

3. The mean of the chi-square distribution is given by the d.f. i.e, E(

and variance is twice the d.f. i.e, v(

= 2v.

4. As v gets larger,

approaches the normal distribution with mean v and

standard deviation .

In practice, it has been determined that the quantity

provides a
better approximation to normality than

itself for values of 30 or more.

The distribution of

has a mean equal to and a standard

deviation equal to 1.

5. The sum of independent

variates is also a

variate. Therefore, if x, is

variate with v, d.f. and

is another

variate with

d.f. indepen-
dent of

, then their sum

is also a

variate with

.


USE OF CHI-SQUARE TABLE:

The table of areas found in the appendix given values of chi-square for various
probabilities and various degrees of freedom.

The value of is given in the column headings and the degrees of freedom, v,
given in the rows and the body of the table given the chi-square value.

The value of chi-square in the appendix table are given for various combinations
of v and 1-.



TESTS OF HYPOTHESIS CONCEINING VARIANCE:

In testing about the variance of a normally distributed population, the
hypothesis is Ho:

where

is some specified value of the population

variance.

We know that

where

is computed from a random sample f size n.

We have dealt with the sampling distributions for sample means and sample
populations.

Now we are dealing with the sampling distribution of sample variances. So for
we have learned that, after certain modification, sample variances follow chi-
distribution.

We begin by considering a random sample of n observations drawn for a
population with unknown mean and unknown variance

. Let us say the

sample is

, -------------------

. The population variance is the expectation

= E [(x-)^2]

Most cases is unknown and

is unknown. So we deal with sample variance

and calculate (

- to compute sample variance.

Confidence interval for the difference between two normal population
means:

DEPENDENT SAMPLES:

For example in pharmaceutical industry say a drug is already in the market,
working well for the people, but a new drug is about to be launched in the
market. What we do is this test the effectiveness of the new drug under similar
condition as the old drug. See whether there is any difference in the result. We
consider samples to be dependent if the values in the other sample. In this
scheme the sample members are chosen in pairs one from each population. Thus
the procedure is often known as method pairs.

INDEPENDENT SAMPLES:

In this scheme samples are drawn independently from two normally distributed
populations with known population variances so that the membership of one of
the sample is not influenced by the membership of the sample.



Hypothesis :-

Hypothesis is an assumption to be tested. In order to reach decisions it is useful to
make assumptions or guesses about the population involved. Such assumptions which
may or may not be true are called statistical hypothesis. They are, in fact, in general
statements about the probability distribution of the population.


Types Of Hypothesis:-

There are mainly two kinds of hypothesis
(1) Null Hypothesis
(2) Alternative Hypothesis


Null Hypothesis:-

A null hypothesis is a claim about a population parameter that is assumed to be true
until it is declared false. Null hypothesis is denoted by H
0.
Is a claim about a
population parameter that will be true if the null hypothesis is false.

Alternative Hypothesis:-

An alternative hypothesis is a claim that will be true if null hypothesis is false.
It is denoted by H
1.
Types of error:- There are two types of error.
(1) Type 1 error
(2) Type 2 error.


Type 1 error:-

Type 1 error occurs when a true null hypothesis is rejected . The value of represents
the probability of committing that kind of error.
= P(H
0
is rejected / H
0
is true)

Type 2 error:-

Type 2 error occurs when false null hypothesis is accepted. It is represented by .
The value of represents the probability of committing that kind of error.
= (H
0
is not rejected / H
0
is false)
The value of 1- is called the test of power. It represents the probability of not making
type 2 error.

Two Tailed Tests:-

A two tailed test has rejection regions in both tails.




Left-tailed Test:-

A left-tailed test has rejection region in left tail.









Right tailed Test:- A right tailed test has rejection region in right tail.



Procedure for Hypothesis testing:

1. set up a hypothesis: the approach to hypothesis testing is not to construct
single hypothesis about the population parameter but rather to set up
two hypothesis. They are:
Null hypothesis, denoted by Ho
Alternative hypothesis, denoted by H1
2. Set up a suitable significance level : the level of significance is generally
specified before any samples are drawn. So that results obtained will not
influence our choice.
3. Determination of a suitable test statistic; the third step is to determine a
suitable test statistic and its distribution. General form of test statistic is
Test statistic = (sample statistic Hypothesized parameter) std. dev of
the statistic
4. Determine the critical region : it is important to specify before the sample
is taken which value of the test statistic will lead to a rejection of Ho and
which lead to acceptance of Ho.
5. Doing computations: we need to perform various computations from a
random sample of size n necessary for the test statistic. Then we need to
see whether sample result falls in the critical region or in the acceptance
region.
6. Making decisions; draw statistical conclusions and decision maker may
take the decision.

Tests of hypothesis concerning large numbers:

Though it is difficult to draw a clear cut line of demarcation between large and
small samples, it is generally agreed that if the size of sample exceeds 30 if
should be regarded as a large sample.

Tests of hypothesis involving large samples are based on the following
assumptions:

1. The sampling distribution of statistic is approximately normal.
2. Values given by the samples are sufficiently close to the population value
and can be used in its place for the standard error of the estimate.
Interpretation of the Probability values or P-values:

The p-value provides more precise information about the strength of the
rejection of the null hypothesis that result from the observed sample mean.
The p-value is the smallest significance level at which Ho can be rejected.
Consider a random sample of size n observations from a population that has a
normal distribution with mean and standard deviation and the resulting
computed sample mean, x-bar.

The Features of the Power Function:

1. The farther the true mean is from the hypothesized mean, the greater is
the power of the test, everything else being equal.
2. The smaller the significance level of the test, the smaller the power,
everything also being equal. Thus reducing the probability of type I error
increases the probability of type II error, but reducing by 0.01 does not
generally increase by 0.01. the change is not linear.
3. The larger the population variance, the lower the power of the test,
everything else being equal.
4. The larger the sample size, the greater the power of the test, everything
else being equal.
5. The power of the test at the critical value equals 0.5 because that a sample
mean is above o=xc is, of course, 0.50.


Maths from the Book

8.2 Two normally distributed populations based on dependent samples of
n=5 observations
a. Find the margin of error for 90% confidence level
Paired T-Test and CI: Before_Ex8.2, After_Ex8.2
Paired T for Before_Ex8.2 - After_Ex8.2
N Mean StDev SE Mean
Before_Ex8.2 5 8.4000 2.6077 1.1662
After_Ex8.2 5 10.2000 3.1145 1.3928
Difference 5 -1.80000 0.83666 0.37417

90% CI for mean difference: (-2.59766, -1.00234)

1, 2
d
n
d
s
ME t
n
o
= =
.83666
2.132
5
ME = = .79772

b. Find the UCL and LCL for a 90% confidence interval
UCL = -2.59766, LCL = -1.00234

c. Find the width of a 95% confidence interval
1, 2
2
d
n
d
s
width ME t
n
o
(
= =
(
(

=
.83666
2 2.776
5
(
(

= 2.07737

8.4 Let X = Without Passive Solar; Y = With Passive Solar;
i i i
d x y =

9,.05
10, 373, 37.3, 1.833
i
n d d t = = = =

( )
2
2806.1/ 9 17.6575
1
i
d
d
d d
s
n

= = =

37.3 1.833(17.6575) / 10
27.0649 47.5351
x y
< <

8.8 Assuming equal population variances, determine the number of degrees
of freedom for each
a. degrees of freedom = 2
x y
n n + = 12 + 14 2 = 24
b. degrees of freedom = 2
x y
n n + = 6 + 7 2 = 11
c. degrees of freedom = 2
x y
n n + = 9 + 12 2 =19
8.10 Assuming unequal population variances, determine the number of
degrees of freedom
a.
2
2 2
1 2
1 2
2 2
2 2
1 2
1 2
1 2
/( 1) /( 1)
s s
n n
v
s s
n n
n n
( | | | |
+
( | |
\ . \ .
=
| | | |
+
| |
\ . \ .
=
2
2 2
6 10
12 14
6 10
/(12 1) /(14 1)
12 14
v
( | | | |
+
| |
(
\ . \ .
=
| | | |
+
| |
\ . \ .
~ 24
b.
2
2 2
1 2
1 2
2 2
2 2
1 2
1 2
1 2
/( 1) /( 1)
s s
n n
v
s s
n n
n n
( | | | |
+
( | |
\ . \ .
=
| | | |
+
| |
\ . \ .
-
2
2 2
30 36
6 10
30 36
/(6 1) /(10 1)
6 10
v
( | | | |
+
| |
(
\ . \ .
=
| | | |
+
| |
\ . \ .
~ 11

c.
2
2 2
1 2
1 2
2 2
2 2
1 2
1 2
1 2
/( 1) /( 1)
s s
n n
v
s s
n n
n n
( | | | |
+
( | |
\ . \ .
=
| | | |
+
| |
\ . \ .
-
2
2 2
16 25
9 12
16 25
/(9 1) /(12 1)
9 12
v
( | | | |
+
| |
(
\ . \ .
=
| | | |
+
| |
\ . \ .
~ 19
d.
2
2 2
1 2
1 2
2 2
2 2
1 2
1 2
1 2
/( 1) /( 1)
s s
n n
v
s s
n n
n n
( | | | |
+
( | |
\ . \ .
=
| | | |
+
| |
\ . \ .
-
2
2 2
30 36
6 7
30 36
/(6 1) /(7 1)
6 7
v
( | | | |
+
| |
(
\ . \ .
=
| | | |
+
| |
\ . \ .
~ 11


8.18 Calculate the margin of error, assuming 95% confidence level
a.
1 1 2 2
2
1 2
(1 ) (1 ) p p p p
ME z
n n
o

= + =
.75(1 .75) .68(1 .68)
1.96
260 200

+
= .083367
b.
1 1 2 2
2
1 2
(1 ) (1 ) p p p p
ME z
n n
o

= + =
.60(1 .60) .68(1 .68)
1.96
400 500

+
= .063062
c.
1 1 2 2
2
1 2
(1 ) (1 ) p p p p
ME z
n n
o

= + =
.20(1 .20) .25(1 .25)
1.96
500 375

+
= .056126

8.20
.01
85 78
120, .7083, 163, .4785, 2.33
120 163
x x y y
x y
n p n p z
n n
= = = = = = = = =


/ 2
(1 ) (1 )
( )
y y
x x
x y
x y
p p
p p
p p z
n n
o

+ =
(.7083)(.2917) (.4785)(.5215)
(.7083 .4785) (2.326)
120 163
+
=
.2298 .132657 = .0971 up to .3625

8.26 How large a sample is needed to estimate the population proportion?
a.
( )
2
2
2
.25 z
n
ME
o
= =
( )
2
2
.25 1.96
.03
= 1067.111. Take a sample of size n =
1068.
b.
( )
2
2
2
.25 z
n
ME
o
= =
( )
2
2
.25 1.96
.05
= 384.16. Take a sample of size n =
385.
c. For ME become half, the sample size must be increased by a larger
proportion
8.28 a.
.05
1.645, .04 z ME = =
2
/ 2
2
.25( ) z
n
ME
o
= =
2
2
(.25)(1.645)
422.8
(.04)
= , take n = 423
b.
2
2
(.25)(1.96)
600.25
(.04)
= , take n = 601
c.
2
2
(.25)(2.33)
542.89
(.05)
= , take n = 543
8.30
.05
1.645, .03 z B = =

2
/ 2
2
.25( ) z
n
ME
o
= =
2
2
(.25)(1.645)
751.7
(.03)
= , take n = 752
8.40 Independent random samples from two normally distributed
populations
a. If the unknown population variances are equal, find a 90%
confidence interval
1 2
2, 2
1 1
( )
n n p
x y
x y t s
n n
o +
+ where
2 2
( 1) ( 1)
2
x x y y
p
x y
n s n s
s
n n
+
=
+


2 2
(10 1)30 (12 1)25
10 12 2
p
s
+
=
+
= 27.3633
1 1
(480 520) 1.725(27.3633)
10 12
+ = -40 20.2106 = -60.21056 to
- 19.7894

b. If the unknown population variances are unequal, find a 90% CI
2
2
( , / 2)
( )
y
x
v
x y
s
s
X Y t
n n
o
+ where
2
2
2
2
2
2
2
/( 1) /( 1)
y
x
x y
y
x
x y
x y
s
s
n n
v
s
s
n n
n n
( | |
| |
+ ( |
|
|
\ . (
\ .
=
| |
| |
+
|
|
|
\ .
\ .
=
2
2 2
2 2
2 2
30 25
10 12
30 25
/(10 1) /(12 1)
10 12
v
( | | | |
+
( | |
\ . \ .
=
| | | |
+
| |
\ . \ .
= 17.606 ~ 18
2 2
30 25
(480 520) 1.734
10 12
+ = -40 20.669 = -60.669 to -19.331

8.42
37,.10
21, 72.1, 11.3, 1.303
x x
n x s t = = = = (df = 37 does not appear
in Appendix Table 7; we used df = 40 to give an approximate answer)
18, 73.8, 10.6
y y
n y s = = =
1 2
2, 2
1 1
( )
n n p
x y
x y t s
n n
o +
+ where
2 2
( 1) ( 1)
2
x x y y
p
x y
n s n s
s
n n
+
=
+


2 2
(21 1)11.3 (18 1)10.6
21 18 2
p
s
+
=
+
= 10.9839
1 1
(72.1 73.8) 1.303(10.9839)
21 18
+ = -1.7 4.5971 = -6.2971 to
2.8971
8.50 Construct a 95% confidence interval of the difference in population
proportions
1
300
.75
400
x
p
n
= = = ,
2
225
.45
500
x
p
n
= = =
1 1 2 2
1 2 2
1 2
(1 ) (1 )
( )
p p p p
p p z
n n
o

+ =
.75(1 .75) .45(1 .45)
(.75 .45) 1.96
400 500

+ = .30 .06085= .23915 up to
.36085

9.4 a. European perspective:
0
:
H
Genetically modified food stuffs are not safe
1
:
H
They are safe
b. U.S. farmer perspective:
0
:
H
Genetically modified food stuffs are safe
1
:
H
They are not safe
9.8 Using the results from the above two exercises, indicate how the critical value
c
x is influenced by sample size. Next indicate how the critical value is
influenced by the population variance.
The critical value
c
x is farther away from the hypothesized value the smaller
the sample size n. This is due to the increase in the standard error with a
smaller sample size.
The critical value
c
x is farther away from the hypothesized value the larger
the population variance. This is due to the increased standard error with a
larger population variance.

9.9 A random sample of n = 25, variance =
2
o and the sample mean is = 70.
Consider the null hypothesis
0
: 80
H

= versus the alternative
1
: 80
H

s . Compute the p-value
a.
2
o = 225.
0
x
z
n

= =
70 80
15 25

= -3.33. ( 3.33)
p
p value P z = < =
.0004
b.
2
o = 900.
0
x
z
n

= =
70 80
30 25

= -1.67. ( 1.67)
p
p value P z = < =
.0475
c.
2
o = 400.
0
x
z
n

= =
70 80
20 25

= -2.50. ( 2.50)
p
p value P z = < =
.0062
d.
2
o = 600.
0
x
z
n

= =
70 80
24.4949 25

= -2.04.
( 2.04)
p
p value P z = < = .0207

9.10 A random sample of n = 25, variance =
2
o and the sample mean is = 70.
Consider the null hypothesis
0
: 80
H

= versus the alternative
1
: 80
H

s . Compute the p-value
e.
2
o = 225.
0
x
z
n

= =
70 80
15 25

= -3.33. ( 3.33)
p
p value P z = < =
.0004
f.
2
o = 900.
0
x
z
n

= =
70 80
30 25

= -1.67. ( 1.67)
p
p value P z = < =
.0475
g.
2
o = 400.
0
x
z
n

= =
70 80
20 25

= -2.50. ( 2.50)
p
p value P z = < =
.0062
h.
2
o = 600.
0
x
z
n

= =
70 80
24.4949 25

= -2.04.
( 2.04)
p
p value P z = < = .0207

9.12
0
: 50
H

> ;
1
: 50
H

< ; reject
0 H
if Z
.10
< -1.28

48.2 50
3 9
Z

= = -1.8, therefore, Reject
0 H
at the 10% level.
9.14 Test
0
: 100
H

s ;
1
: 100
H

> , using n = 25 and alpha = .05
a. 106, 15 x s = = . Reject if
0
1, 2 n
x
t
s n
o

> ,
106 100
15 25

= 2.00. Since 2.00 is

greater than the critical value of 1.711, there is sufficient evidence to reject
the null hypothesis.
b. 104, 10 x s = = . Reject if
0
1, 2 n
x
t
s n
o

> ,
104 100
10 25

= 2.00. Since 2.00 is

greater than the critical value of 1.711, there is sufficient evidence to reject
the null hypothesis.
c. Assuming a one-tailed lower tailed test, 95, 10 x s = = . Reject if
0
1, 2 n
x
t
s n
o

> ,
95 100
10 25

= -2.50. Since -2.50 is less than the critical

value of -1.711, there is sufficient evidence to reject the null hypothesis.
d. Assuming a one-tailed lower test, 92, 18 x s = = . Reject if
0
1, 2 n
x
t
s n
o

> ,
92 100
18 25

= -2.22. Since -2.22 is less than the critical value of -1.711,

there is sufficient evidence to reject the null hypothesis.


9.20
0 1
: 0; : 0;
H H

= <
2.91 0
11.33 170
t

= = -3.35, p-value is < .005. Reject
0 H
at any common level
of alpha.

9.24
0 1
: 20; : 20;
H H

= = reject
0 H
if |t
8, .05/2
| > 2.306

20.3556 20
.6126 9
t

= = 1.741, therefore, do not reject
0 H
at the 5% level


9.28 A random sample is obtained to test the null hypothesis of the proportion of
women who said yes to a new shoe model.
0 1
: .25; : .25;
p p
H H
s > .
What value of the sample proportion is required to reject the null hypothesis
with alpha = .03?
a. n = 400. Reject
0 H
if
0 0 0
(1 ) /
c
p p p z p p n
o
> = + = .25 +1.88
(.25)(1 .25) / 400 = .2907
b. n = 225. Reject
0 H
if
0 0 0
(1 ) /
c
p p p z p p n
o
> = + = .25 +1.88
(.25)(1 .25) / 225 = .30427
c. n = 625. Reject
0 H
if
0 0 0
(1 ) /
c
p p p z p p n
o
> = + = .25 +1.88
(.25)(1 .25) / 625 = .28256
d. n = 900. Reject
0 H
if
0 0 0
(1 ) /
c
p p p z p p n
o
> = + = .25 +1.88
(.25)(1 .25) / 900 = .2771

9.34
0 1
: .5; : .5; p p
H H
= >

.56 .5
(.5)(.5) / 50
z

= = .85, p-value = 1 F
Z
(.85) = 1 .8023 = .1977
Therefore, reject
0 H
at alpha levels in excess of 19.77%
9.38 What is the probability of Type II error if the actual proportion is
a. .52 P = .
*
(.46 .54| ) P p p p | = s s = =
* *
* * * *
.46 .54
(1 ) (1 )
p p
P z
p p p p
n n
(
(

(
s s
(

(


=
.46 .52 .54 .52
.52(1 .52) .52(1 .52)
600 600
P z
(
(

( s s
(
(

= P(-2.94 z .98) = .4984 + .3365 =
.8349
b. .58 P = .
*
(.46 .54| ) P p p p | = s s = =
.46 .58 .54 .58
.58(1 .58) .58(1 .58)
600 600
P z
(
(

( s s
(
(


= P(-5.96 z -1.99) = .5000 .4767 = .0233
c. .53 P = .
*
(.46 .54| ) P p p p | = s s = =
.46 .53 .54 .53
.53(1 .53) .53(1 .53)
600 600
P z
(
(

( s s
(
(


= P(-3.44 z .49) = .4997 + .1879 = .6876
d. .48 P = .
*
(.46 .54| ) P p p p | = s s = =
.46 .48 .54 .48
.48(1 .48) .48(1 .48)
600 600
P z
(
(

( s s
(
(


= P(-.98 z 2.94) = .3365 + .4984 = .8349
e. .43 P = .
*
(.46 .54| ) P p p p | = s s = =
.46 .43 .54 .43
.43(1 .43) .43(1 .43)
600 600
P z
(
(

( s s
(
(


= P(1.48 z 5.44) = .5000 .4306 = .0694



9.40 a.
0 H
is rejected when
3
.4 64
X
> 1.645 or when X > 3.082. Since the
sample mean is 3.07% which is less than the critical value, the decision is do not
reject the null hypothesis.
b. The | = P(Z <
3.082 3.1
.4 64

) = 1 F
Z
(.36) = .3594. Power of the test = 1 - |
= .6406

9.48
2 2
0 1
: 500; : 500;
H H
o o s > reject
0 H
if
2
(7,.10) _ > 12.02

2
2
2
( 1) 7(933.982)
500
n s
_
o

= = = 13.0757, Therefore, reject

0 H
at the 10%
level

9.50
2 2
0 1
: 300; : 300;
H H
o o = =

2
29(480)
300
_ = = 46.4, p-value = .0214. Reject
0 H
at the 5% level

7.42 a.
2 2 2
20,.975 20,.025
21, 16, .05, 9.59, 34.17 n s taking o _ _ = = = = =

2 2
2
2 2
1, / 2 1,1 / 2
( 1) ( 1)
n n
n s n s
o o
o
_ _

< < =
2
21(16) 21(16)
34.17 9.59
o < < = 9.8332 <
2
o <
35.036

b.
2 2
15,.975 15,.025
16, 8, 6.26, 27.49 n s _ _ = = = =

2 2
2
2 2
1, / 2 1,1 / 2
( 1) ( 1)
n n
n s n s
o o
o
_ _

< < =
2 2
2
15(8) 15(8)
27.49 6.26
o < < = 34.9218 <
2
o <
153.3546

c.
2 2
27,.995 27,.005
28, 15, 11.81, 49.64 n s _ _ = = = =

2 2
2
2 2
1, / 2 1,1 / 2
( 1) ( 1)
n n
n s n s
o o
o
_ _

< < =
2 2
2
28(15) 28(15)
49.64 11.81
o < < = 126.9138 <
2
o
< 533.446


7.48
2 2 2
17,.05 17,.95 18, 108.16, 27.59, 8.67 n s _ _ = = = =

2 2
2
2 2
1, / 2 1,1 / 2
( 1) ( 1)
n n
n s n s
o o
o
_ _

< < =
2
17(108.16) 17(108.16)
27.59 8.67
o < <
= 66.6444 up to 212.0784. Assume that the population is normally
distributed.
9.56 a. False. The significance level is the probability of making a Type I error
falsely rejecting the null hypothesis when in fact the null is true.
b. True
c. True
d. False. The power of the test is the ability of the test to correctly reject a
false null hypothesis.
e. False. The rejection region is farther away from the hypothesized value at
the 1% level than it is at the 5% level. Therefore, it is still possible to reject at
the 5% level but not at the 1% level.
f. True
g. False. The p-value tells the strength of the evidence against the null
hypothesis.

9.58 a.
776 800
(
120 100)
P Z o

= < ) = P(Z < -2) = .0228
b.
776 740
( )
120 100
P Z |

= > = P(Z > 3) = .0014
c. i) smaller ii) smaller
d. i) smaller ii) larger





9.60
0 1
: .5; : .5; p p
H H
= =

.4808 .5
(.5)(.5) /104
z

= = -.39, p-value = 2[1-F
Z
(.39)] = 2[1-.6517] = .6966
Therefore, reject
0 H
at levels in excess of 69.66%