FHMM 1324

Mathematics For Business 2

Chapter 4
Estimation and Confidence Interval

1
 Topics
Distribution of Sample Means and Sample
Proportions

Point Estimates of population Proportion

and Mean from Large Sample or Population
with known Variance

Choosing Appropriate Sample Size

2
 Introduction
A sample is a subset of a population

Population Sample

3
 Introduction
Parameters Numerical measures obtained from a population to describe the
characteristics of the population.

Statistic Numerical measures obtained from a sample to describe the

characteristics of a sample.

Example :
Mean height of all students in Malaysia is 1.48 m.
Population parameter = the mean height of all the students
= 1.48 m.

Mean height of students in Utar is 1.42 m.

Sample statistic = the mean height of students in Utar
= 1.42m

4
 Introduction
The symbols and terms used in the sampling theory

Numerical measure Parameter Statistic

Size N n
Mean x
Variance 2 s2
Proportion p

5
Sampling From An Infinite Population
(Or Finite Population With Replacement)
All possible random samples of size n with values
x1,, x2, x3 ,.,xn are selected from an infinite
population ( or a finite population with
replacement) with mean and variance and
2

their means are calculated.

6
Sampling From An Infinite Population
(or Finite Population With Replacement)

A probability distribution of all these

possible sample means is called the
sampling distribution of the sample
mean.

7
 Sampling distribution of
the sample mean
The mean of the sample means:

x E( x )
The variance of the sample means:

2
x 2
Var ( x )
n
8
 Standard Error Of The Mean
The standard deviation of the sampling distribution
of the sample mean is also known as the standard
error of the mean, denoted by x .

x
n
n sample size
population mean
population standard deviation 9
 Standard Error Of The Mean
If the sample size n is large, then the standard
error of the mean is small.
Hence, the sample means will be closer to the
population mean.

10
 Standard Error of The mean

2
x E ( X ) E ( X )
2

where

E( X ) x P( X x )

E ( X ) x P( X x )
2 2
11
 Example 4.1
Consider a sample of size 2 taken from a set of
integers {0, 2, 4, 6, 8} with replacement. The total
number of samples is 5 x 5 = 25 and is listed below.

(0,0), (2,0), (4,0), (6,0), (8,0),

(0,2), (2,2), (4,2), (6,2), (8,2),
(0,4), (2,4), (4,4), (6,4), (8,4),
(0,6), (2,6), (4,6), (6,6), (8,6),
(0,8), (2,8), (4,8), (6,8), (8,8).

Find the mean and standard deviation for the sample

means. Hence, find the probability distribution of the
12
sample means and illustrate it with a graph.
Example 4.1 Solution

13
Example 4.1 Solution

14
Example 4.1 Solution

15
 Example 4.1 Solution

2
E( X )
x P( X x )
2

0 (0.04) 1 (0.08) 2 (0.12) 3 (0.16) 4 (0.20)

2 2 2 2 2

5 (0.16) 6 (0.12) 7 (0.08) 8 (0.04)

2 2 2 2

20

16
 Example 4.1 Solution

2
x E ( X ) E ( X )
2

20 4 2

2
17
Example 4.1 Solution

18
Example 4.2

19
 Example 4.2 Solution

E( X ) 4
2 8 2
4 Var ( X ) 4
n 2 n

20
 Example 4.3

Ans: 63.25, 20

21
 Example 4.3 Solution

Standard error of the mean

=
x
n
For sample A: n = 10

200
For sample B:

n =100x
63.25
10

200
x 20
100
22
 Example 4.3 Solution

When sample size increases from 10 to 100, the

standard error of the mean decreases from
63.25 to 20.

23
 Central Limit Theorem
(a) Sampling distribution of the mean obtained
from any population.

(b) Sampling distribution of the mean obtained

from a normal population.

(c) Proportion

(d) Sampling distribution of the proportion.

24
(a) Sampling Distribution Of The
Mean From Any Population
f (x ) f (x ) f (x )

xx x xx

f (x ) f (x ) f (x )

x x x
x
25
 (a) Sampling Distribution Of The
Means From Any Population
If samples of size n are selected from any
population (with mean and variance 2), the
sampling distribution of the sample mean is
approximately a normal distribution with mean
2
and variance provided n is large (n 30).
n

26
(a) Sampling Distribution Of The
Mean From Any Population

2

X ~ N ( , ) n 30
n

27
Example 4.4

28
Example 4.4 (a) Solution

29
Example 4.4 (a) Solution

30
Example 4.4 (a) Solution

31
Example 4.4 (b) Solution

32
Example 4.4 (b) Solution

33
 Example 4.5
Find the probability that the sample mean
selected at random for a sample of size 30 taken
from a population with binomial distribution,
parameters n = 10 and p = 0.5 will exceed the
value of 5.5.

34
Example 4.5 Solution

35
 Example 4.6
A random sample of size m is selected form a
population with binomial distribution,
parameters n = 20 and p = 0.4. It is found that
4% of the sample mean is less than 7.5. Find
the value of m.

Ans: m = 59
36
 Example 4.6 Solution

X : B(20, 0.4)
E ( X ) np 20(0.4) 8
Var ( X ) npq 20(0.4)(0.6) 4.8)
4.8
X ~ N (8, )
m
P ( X 7.5) 0.04
37
 Example 4.6 Solution

4.8
P(Z
7.5 8
) 0.04
0.2857
4.8 m
m m 58.80
P ( Z 1.75) 0.04
59

7.5 8
1.75
4.8

m

38
 (b) Sampling Distribution of The
Mean From A Normal Population
f (x ) f (x ) f (x )

x x x

The diagram 4.4 (a) show a population with a normal

distribution. Then the sampling distribution of the mean
will also be normally distributed for any sample size as
shown in diagrams (b), (c ) and (d). 39
 (b) Sampling Distribution of The
Mean From A Normal Population
If the samples are selected randomly from a normal
population (with mean and variance 2), the sampling
distribution of the sample means will also be normally
distributed with mean and variance 2 for any
n
sample size

2

X ~ N ( , )
n
40
 Example 4.7
The length of a type of worm is normally distributed
with mean 15cm and standard deviation cm.
30
Find the probability that the length of a worm
selected at random lies between 14cm and 16cm.

A sample of size 25 worms is chosen randomly and

the mean length is calculated. Find the probability
that the mean length lies between 14cm and 16cm.

Ans: 0.1428; 0.6372 41

 Example 4.7 Solution

X ~ N (15,30)
P (14 X 16)
P ( 143015 Z 163015 )
P (0.18 Z 0.18)
1 P ( Z 0.18) P ( Z 0.18)
1 2 P ( Z 0.18)
1 2(0.4286)
0.1428
42
 Example 4.7 Solution

X ~ N (15, 30
25 )

P(14 X 16)
P ( 143015 Z 163015 )
25 25

P (0.91 Z 0.91)
1 P ( Z 0.91) P ( Z 0.91)
1 2 P ( Z 0.91)
1 2(0.1814)
0.6372 43
 Example 4.8
A sample of size n is taken from a normally
distributed population with mean 80 and
variance 36. Find the value of n if the
probability of the sample mean exceeds 78 is
0.975.

Ans: n = 35
44
 Example 4.8 Solution

X ~ N (80, 36n ) P ( Z 1.96) 0.025

P ( X 78) 0.975 1.96

7880
36
n

P(Z 78 80
) 0.975
36
n
36
n 1.020

1 P( Z 78 80
36
n
) 0.975 n 34.6
n 35
P(Z ) 0.025
78 80
36
n

45
 Example 4.9
A large random sample of size n is taken from a
normal population with mean 80 and standard
deviation 5. Find the least value of n if the sample
mean

(a) exceeds the population mean by at least 2

with a probability of less than 0.01,

(b) differs from the population mean by at least 2

with a probability of less than 0.01.

Ans: (a) n = 34; (b) n = 42 46

 Example 4.9 Solution

(a) (b) P( X 80 2) 0.01

52
X ~ N (80, n ) P( X 82 or X 78) 0.01
P( X 82) P( X 78) 0.01
P( X 80 2) 0.01
P( Z 825 80 ) P( Z 78 80
5 ) 0.01
P( Z 825 80 ) 0.01 n n

n 2 n 2 n
P( Z ) P( Z ) 0.01
5 5
P( Z 2 n
5 ) 0.01 2 P ( Z 2 5 n ) 0.01
2 n
5 2.33 P( Z 2 n
5 ) 0.005
2 n
2.58
n 33.9 5

n 41.6
Least value of n is 34. Least value of n is 42.
47
(c) Proportion
A proportion is a comparison of two values.
If k is the number of elements with a particular
characteristic taken from a set of n elements, then the
proportion of elements with the particular feature is .

k
denotes the proportion from a population with a
n
particular characteristic.

P denotes the proportion from a sample with a particular
characteristic.

48
 (c) Sampling Distribution
of The Proportion
If is the proportion of a population with a
particular characteristic and random samples of size n
are selected, then the probability distribution of the
sample proportions p that have the particular
characteristic is approximately normal with mean
and variance (1 , ) provided n is sufficiently large.
n

(1 )
p ~ N ,
n 49
 (c) Sampling Distribution
of The Proportion
A probability distribution of all sample
proportions with a particular characteristic from
a population is called the sampling distribution
of the proportion.
The mean of the sampling distribution of the
proportion is given by:
p E ( p)
50
 (c) Standard Error For The Sampling
Distribution Of The Sample Proportion
The standard deviation of the sampling distribution of
the sample proportion is denoted by p .

(1 )
p
n
It is also known as the standard error of the
proportion.
If the size n is large , then the standard error will be
small. Thus, the sample proportion will be closer to
the population proportion.
51
 Example 4.10
A company manufactures a type of light bulbs. It
is known that 5% of the bulbs are defective. Find
the probability that a sample of 400 light bulbs
chosen at random will yield a proportion of
defective bulbs of

(a) at least 4% (b) at most 5.5%

(c) more than 4.5% (d) less than 4.8%.

Ans: (a) 0.8212; (b) 0.6772; (c) 0.6772; (d) 0.3821 52

 Example 4.10 Solution

(a) 0.05(0.95)
p ~ N (0.05, 400 )
~ N (0.05, 0.00011875)
P( p 0.04)
P( Z 0.00011875
0.04 0.05
)
P( Z 0.92)
1 P( Z 0.92)
1 0.1788
0.8212
53
 Example 4.10 Solution

(b) p ~ N (0.05, 0.05(0.95)

400 )
~ N (0.05, 0.00011875)
P( p 0.055)
P( Z 0.055 0.05
0.00011875
)
P( Z 0.46)
1 P( Z 0.46)
1 0.3228
0.6772
54
 Example 4.10 Solution

(c) 0.05(0.95)
p ~ N (0.05, 400 )
~ N (0.05, 0.00011875)
P( p 0.045)
P(Z 0.045 0.05
0.00011875
)
P ( Z 0.46)
1 P( Z 0.46)
1 0.3228
0.6772
55
 Example 4.10 Solution

(d) p ~ N (0.05, 0.05(0.95) )

400

~ N (0.05, 0.00011875)
P( p 0.048)
P ( Z 0.048 0.05
0.00011875
)
P ( Z 0.18)
P ( Z 0.18)
0.4286

56
 Example 4.11
The proportion of students of a secondary school
who wear glasses is 0.35. A study is taken to
determine the proportion of students who wear
glasses by choosing a sample A of 30 students
and another sample B of 50 students. Find the
standard error of proportion for these two samples
and comment on your results.

Ans: SA = 0.0871; SB = 0.0675 57

 Example 4.11 Solution

When n A 30
0.35(1 0.35)
Standard error of sample A: p 0.0871
30

When nB 50
0.35(1 0.35)
Standard error of sample B: p 0.0675
50

58
 Example 4.11 Solution

When sample size increase from 30 to 50, the

standard error of the sample proportion of
student wearing glasses decrease from 0.0871 to
0.0675.

Hence, the larger the sample size the smaller

the standard error.

59
 Point Estimation
Point estimation refers to the procedure whereby a value of
sample statistic is computed from the data of a sample to be
used as an estimate of the population parameter.

The sample statistic used to estimate a population

parameter is called a point estimator.

Population parameter Point estimator

x
S
P 60
 Unbiased Estimator
A point estimator is unbiased if the
mean of its sampling distribution
equals to the value of the parameter
of interest.

An unbiased point estimator does not

systematically over estimate or under
estimate the target parameter.
61
 Unbiased Estimator
A Statistic is an unbiased estimator of the
population parameter if the expected value
of the statistic is .

That is, if U is the unbiased estimator of ,

then E (U )
Example:
E ( x)
x is an unbiased estimator of
62
 Properties of Point Estimators
Of all the unbiased estimators, we prefer the
estimator whose sampling distribution has the
smallest spread or variability.

63
 Unbiased Estimator For The
Population Mean
If X1, X2, X3 , Xn is a sample of size n taken
from a population with mean and variance 2,
then the unbiased estimator for the population mean
is
n
1 1
X X i ( X 1 X 2 .... X n )
n i 1 n
64
 Unbiased Estimator For The
Population Variance
If X1, X2, X3 , Xn is a sample of size n taken
from a population with mean and variance 2,
then the unbiased estimator for the population
variance 2 is
n

(X X ) 2

S
2 i 1
n 1
65
 Unbiased Estimator For The
Population Variance
The unbiased estimator formulae for the population
variance can have other forms as shown below :

( x x ) 2
x 2

( x ) 2

(a) S 2 (b) S
2 n
n 1 (n 1)

(c ) S 2
x 2
n ( x ) 2

n 1

For grouped data, replace n with f in the formulae above.

66
 Example 4.12
Find the unbiased estimator for the population
mean and variance from the following sample
data:
( a ) 46, 48,51, 50, 45, 53, 50, 48
(b)
x 120, x 2 2102, n 8

(c ) x 128, ( x x ) 312,
2
n8
(d ) x 20 21 22 23 24 25
f 4 14 17 26 20 9
Ans: (a) 48.88, 6.9821; (b) 15, 43.1429; (c) 16; 44.5714;
67
(d) 22.7889, 1.8089
 Example4.12
Example 4.15 Solution

(a) Unbiased estimator for population means is

X 18 (46 48 51 50 45 53 50 48)
48.875
Unbiased
n
estimator for population variance is:
X
i 1
2
46 2 482 512 452 532 50 2 482

19159
n

X i
2
n( X ) 2
S 2
i 1

n 1
19159 8(48.875) 2

8 1 68
6.9821
 Example4.12
Example 4.15 Solution

(b) Unbiased estimator for

population means is

X
x Unbiased estimator for
n population variance is

120
S
2 x 2
n ( X ) 2

8 n 1
15
2102 8(15) 2

8 1
43.1429
69
 Example4.12
Example 4.15 Solution

(c) Unbiased estimator for

population means is

X
x Unbiased estimator for
n population variance is
n
128

8
i
( X X ) 2

S2 i 1

16 n 1
312

8 1
44.5714
70
 Example4.12
Example 4.15 Solution

(d) Unbiased estimator for

population means is

fx Unbiased estimator for

X population variance is
n
2051

90
22.7889
20512
46901
S2 90
90 1
1.8089

71
 Unbiased Estimator For The
Population Proportion
In a random sample of size n taken from a
population with a specific character, then the
unbiased estimator for the population proportion
with the specific character p
is k where k is
n
the number of the sample elements with the
specific character.

72
 Example 4.13
A research is conducted in a town to find the
number of residents who owns a credit card. Out
of a random sample of 120 residents, 25 of them
own a credit card. What is the unbiased estimator
for the proportion of town residents who owns a
credit card?

Ans: 0.2083 73
 Example 4.13 Solution

The sample size, n=120

The number of residents who owns a credit card,
k=25
Therefore, the unbiased estimator for the proportion
of town residents who owns a credit card is

25
p 0.2083
120
74
 Interval Estimation
An interval is constructed around the point estimate so
that the interval is likely to contain the parameter at a
specified probability denoted by (1 ).
The interval is then said to be constructed with respect to
a given confidence level denoted by (1 )100%. The
interval established is called the confidence interval.
Usually,
Usually,1-
1-
Suppose 1- = .95 and that
the estimator has a normal
distribution, then 95% of
the values of the estimator
lie in the interval: Parameter
Parameter1.96SE
1.96SE
75
 To Change the Confidence Level
To change to a general confidence level, 1-, pick a
value of z that put the area of 1- in the center of the z
distribution. (Refer to Table 4 Percentage Points of
the Normal Distribution for the za/2 values)
Tail area za/2
0.05 1.6449
2 2 0.025 1.9600
0.01 2.3263
0.005 2.5758

100(1-)%
100(1-)% Confidence
Confidence Interval: Estimator zza/2a/2SE
Interval: Estimator SE
76
 Interval Estimation
(a) Confidence interval for the population mean
with known variance 2.

(b) Confidence interval for the population mean

with unknown variance 2.

(c) Confidence interval for the population

proportion

77
 Interval Estimation
(d) Interpretation of confidence interval

(e) Sample size to estimate the population

mean when the standard deviation is
known

(f) Sample size to estimate the population

proportion

78

(1-(w
)x1h-
0
%
f C
o
n
ierz)202%d
iS
tC
x e
zo(nX
f
Mean

c
ind3
With known variance,

With unknown variance,

2n e
r v a l f
no r 30p ula
t
i
o
e01c)2in(tuenrvbaialsfeodr:tim
Confidence Interval For The Population

79
n m
e
a

:
n
aor)
(1- onfidpecz2inteprv1alforpnu3l0ationprotin:
)0%
C
Confidence Interval for the
Population Proportion

80
 Example 4.14
A random sample of n = 50 males showed a mean
average daily intake of dairy products equal to 756
grams with a standard deviation of 35 grams. Find
a 95% confidence interval for the population
average .

Ans: (746.2985 , 765.7015 )

81
 Example 4.14 Solution

(1 )100% 95%
1 0.95
0.05

0.025
2
Z 0.025 1.96

82
 Example 4.14 Solution

95% confidence interval

for : S
x 1.96
n
35
756 1.96
50
756 9.70
(746.2985, 765.7015) grams
83
 Example 4.15
Refer to the Example 4.15, find a 99% confidence
interval for the population average daily intake
of dairy products for men.

Note:
The interval must be wider to provide for the increased
confidence that does indeed enclose the true value of m.

Ans:( 743.2504 ,768.7496)

84
 Example 4.15 Solution

(1 )100% 99%
1 0.99
0.01

0.005
2
Z 0.005 2.5758

85
 Example 4.15 Solution

99% confidence interval

for : S
x 2.5758
n
35
756 2.5758
50
756 12.75
(743.2504,768.7496) grams 86
Ans:(0.65,781) Example 4.16
Of a random sample of n = 150 college students,
104 of the students said that they had played on a
soccer team during their K-12 years. Estimate the
proportion of college students who played soccer in
their youth with a 98% confidence interval.

87
 Example 4.16 Solution

k (1 )100% 98%
P
n 1 0.98
104 0.02
p
150
0.01
2
Z 0.01 2.3263
88
(p2.361n)
=
104
150
p Example 4.16
98% confidence interval
for :

2.3263

0.6933 0.0876
104
150

(0.6057, 0.7809)
(1
150

104
150
)
Solution

89
Ans:(8.70,1) Example 4.17
Random samples of 10 fishes are caught from a
pond and its length, in cm, were measured. The
results are as follows: 9.1, 9.1, 11.3, 10.7, 9.8,
10.2, 10.1, 9.7, 9.9 and 9.5. The lengths of fishes
in the pond are normally distributed with variance
4. Find a 95% symmetrical confidence interval
for the mean lengths of all fishes in the pond.

90
 Example 4.17 Solution

(d) Unbiased estimator for (1 )100% 95%

population means is
1 0.95
X
x
0.05
n


99.4 0.025
10 2
9.94 Z 0.025 1.96

91
 :
x1.96n Example 4.17
95% confidence interval
for

= 9.94 1.96

9.94 1.24
2
10

(8.7004,11.1796)

Solution

92
 Example 4.18

Ans:(90.0062,100.9938)
93
(1 )100% 98%
1 0.98
0.02

2
0.01

Z 0.01 2.3263
x=
Example 4.18

for :

2 .36386,x2.36
8
Solution

98% confidence interval

94
xx2.32.366881,0x2.2.36.(12)8.2,108
Example 4.18 Solution

95
 Example 4.18 Solution

25.8769 and x 95.50

(1 )100% 92% 92% Confidence Interval for :
1 0.92 25.8769
95.50 1.7507
0.08 68
95.50 5.4938
0.04 (90.0062,100.9938)
2
Z 0.04 1.7507
96
 Example 4.19
A random sample of size 100 is taken from a
normal population and the data is summarized
below:
x 1000 and ( x x ) 2
170.8
(a)Find the unbiased estimator of the population
mean, .
(b)Find a 99% confidence interval for the
population mean, .
Ans: (9.6617, 10.3383) 97
 2
Example 4.19
For a 99% confidence interval:
(1 )100% 99%
1 0.99
0.01

0.005
=

S
2

2

(
n

x
n 1
10

x ) 2
z20.152.170
From Table 4 (Percentage Points
of the Normal Distribution),
z z0.005 2.5758

(a)The unbiased estimator of

x 1000
100

170.8
Solution

100 1
1.7253

98
S1.72531 Example 4.19

(b) 99% Confidence Interval for is

10 2.5758
1.3135

10 0.3383
100

(9.6617,10.3383)
Solution

99
 Example 4.20
A supervisor wants to know the proportion of
defective electronic components produced by a
machine. He inspects a random sample of 200
electronic components and found that 40
electronic components is defective. Find a
98% symmetrical confidence interval for the
proportion of all electronic components produced
by the machine.

100
Ans: p = 0.2; (b) (0.1342, 0.2658)
 Example 4.20 Solution

The sample proportion of defective electronic

components is p k 40 0.2
n 200
For a 98% symmetrical confidence interval;
(1 )100% 98%
1 0.98
0.02

0.01
2
Z 0.01 2.3263 101
 Example 4.20 Solution

98% symmetrical confidence interval for :

(0.2)(0.8)
0.2 2.3263
200
0.2 0.0658
(0.1342, 0.2658)

102
 Interpretation Of Confidential Interval
Consider the IQ scores of a normal population with
mean = 4.5 and standard deviation = 2.87.
In a study of sampling distribution of the IQ scores,
the mean score of a randomly chosen sample of 40
participants x is found to be 3.975. Thus, a 90%
confidence interval for the mean score is

2.87
x Z 2 3.975 (1.6449
n 40
(3.975 0.7464)
(3.2286, 4.7214) 103
Interpretation Of Confidential Interval
Therefore, we are 90% confident that the
interval (3.2286, 4.7214) will contain the
population mean .

lies in this interval

3.2286 4.7214

104
Interpretation Of Confidential Interval

If we perform the sampling distribution with another

10 randomly selected samples of size 40 each, 10
confidence intervals will be obtained.

Hence for 90% confidence interval, we can expect

0.9 x 10 = 9 of these confidence intervals to contain
the mean population = 4.5.

105
Interpretation Of Confidential Interval
Table below shows the 10 samples and the confidence
intervals obtained in a sampling study.
Sample Sample mean 90% Confidence Interval
1 4.64 (3.89, 5.39)
2 4.56 (3.81, 5.31)
3 3.96 (3.21, 4.71)
4 5.12 (4.37, 5.87)
5 4.24 (3.49, 4.99)
6 3.44 (2.69, 4.19)
7 4.60 (3.85, 5.35)
8 4.08 (3.33, 4.83)
9 5.20 (4.45, 5.95)
10 4.88 (4.13, 4.63) 106
 Interpretation Of Confidential Interval
Figure below is a line segment representation of the
confidence intervals.

Notice that 9 out of 10 confidence intervals contains

the mean = 4.5, i.e. the sixth interval (2.69 , 4.19) do
107
not contain .
 Choosing Appropriate Sample Size
(i) Determining Sample size to estimate the finite population
mean when the standard deviation is known.

2

n Z
E
2
where the estimation error

E Z
2 n
108
 Example 4.21
The income of a fresh graduate of a university is
known to be normally distributed with standard
deviation RM1000. Find the least sample size
that must be taken so that mean income is
estimated with an error within RM500 with a
confidence of 95%.

Ans: n = 16

109
 Example 4.21 Solution

For 95% confidence interval , 0.05 z z 0.025 1.96

2
Estimator error, E= 500 ; standard deviation, =1000.

2
2
1000
hence, n z 1.96 15.3664
E 500
2
Therefore, the minimum sample size that should be taken is 16 (round
to the next integer).
110
 Choosing Appropriate Sample Size
(ii) Sample size to estimate the population proportion.
If an estimate of the population proportion is not
known, use p = .
2
z

n 2 p (1 p )
E


where the estimation error
p (1 p )
Ez
n
2 111
 Example 4.22
A study is conducted in a particular school to
determine the proportion of pupils who smokes.
It is found that 8 out of 120 pupils selected at
random smoke. Find the least sample size so
that the difference between the sample
proportion and the population proportion is less
than 0.05 with confidence at 99%.

Ans: n = 166
112
 Example 4.22 Solution

From the initial study conducted, it is found that the proportion of

8
pupils who smokes is p 0.0667
120

For 99% confidence interval , =0.01

z z0.005 2.5758

2

The estimation error, E =0.05

113
 Example 4.22 Solution

2
z

The sample size , n 2 p(1 p)
r


2
2.5758
(0.0667)(1 0.0667)
0.05
165.2081 166

114
 Example 4.23
An IT company in Johor is interested to know
the proportion of students in the state who own
a computer. Find the least sample size so that
the difference between the sample proportion
and the population proportion is less than 0.1
with a confidence interval of 98%.

Ans: n = 136
115
 Example 4.23 Solution

No information about the proportion of students in Johor who

own a computer. Use p = .

.
For 98% confidence interval , = 0.02

z z0.01 2.3263

2

116
 Example 4.23 Solution

2
z

The sample size, n 2
2 E


2
2.3263

2 0.1
135.2918 136

117
118