Slides Prepared by

JOHN S. LOUCKS
St. Edward’s University

© 2003 South-Western /Thomson Learning™

1
 Chapter 8
Interval Estimation
 Interval Estimation of a Population Mean:
Large-Sample Case
 Interval Estimation of a Population Mean:
Small-Sample Case
 Determining the Sample Size to Estimate 
 Interval Estimation of a Population Proportion

x

[--------------------- x ---------------------]
[--------------------- x ---------------------]
[--------------------- x ---------------------]

© 2003 South-Western /Thomson Learning™

2
 Interval Estimation of a Population Mean:
Large-Sample Case
 Margin of Error and the Interval Estimate
 Probability Statements about the Sampling
Error
 Constructing an Interval Estimate:
Large-Sample Case with Assumed Known
 Calculating an Interval Estimate:
Large-Sample Case with Estimated by s

© 2003 South-Western /Thomson Learning™

3
 Margin of Error and the Interval Estimate

 A point estimator does not provide information

about how close the estimate is to the
population parameter.
 For this reason statisticians usually prefer to
construct an interval estimate.
 An interval estimate is constructed by
subtracting and adding a value, called the
margin Point
of error , to a point estimate.
Estimate  Margin of Error

© 2003 South-Western /Thomson Learning™

4
 Margin of Error and the Interval Estimate

 An interval estimate of the population mean is:

x Margin of Error

 An interval estimate of the population

proportion is:
p Margin of Error

© 2003 South-Western /Thomson Learning™

5
 Sampling Error

 The absolute value of the difference between

an unbiased point estimate and the population
parameter it estimates is called the sampling
error.
 For the case of a sample mean estimating a
population mean:
Sampling Error = | x  |

© 2003 South-Western /Thomson Learning™

6
 Probability Statements
About the Sampling Error
 Knowledge of the sampling distributionx of
enables us to make probability statements
about the sampling error even though the
population mean  is not known.
 A probability statement about the sampling
error is a precision statement.

© 2003 South-Western /Thomson Learning™

7
 Probability Statements
About the Sampling Error
 Precision Statement
There is a 1 -  probability that the
value of a sample mean will provide a
z /2  x error of
sampling or less.
Sampling
distribution
of x
1 -  of all
/2 x values /2

x


© 2003 South-Western /Thomson Learning™

8
 Interval Estimate of a Population Mean:
Large-Sample Case (n > 30)
 With Assumed Known

x z /2
n
where: is the sample mean
x
1 - is the confidence coefficient
z/2 is the z value providing an area of
/2 in the upper tail of the standard
normal probability distribution
 is the population standard deviation
n is the sample size

© 2003 South-Western /Thomson Learning™

9
 Interval Estimate of a Population Mean:
Large-Sample Case (n > 30)
 With Estimated by s
In most applications the value of the
population standard deviation is unknown. We
simply use the value of the sample standard
deviation, s, as the point estimate of the
population standard deviation.
s
x z /2
n

© 2003 South-Western /Thomson Learning™

10
 Example: National Discount, Inc.

National Discount has 260 retail outlets

throughout the United States. National
evaluates each potential location for a new retail
outlet in part on the mean annual income of the
individuals in the marketing area of the new
location.
Sampling can be used to develop an interval
estimate of the mean annual income for
individuals in a potential marketing area for
National Discount.
x
A sample of size n = 36 was taken. The sample
mean, , is $21,100 and the sample standard
deviation, s, is $4,500. We will use .95 as the
confidence coefficient in our interval estimate.
© 2003 South-Western /Thomson Learning™
11
 Example: National Discount, Inc.

 Precision Statement
There is a .95 probability that the value
of a sample mean for National Discount will
provide a sampling error of $1,470 or less…….
determined as follows:
95% of the sample means that can be
x
observed are within + 1.96 of the
population
s mean
4 ,500 .
x   750 x
n 36
If , then 1.96 =
1,470.

© 2003 South-Western /Thomson Learning™

12
 Example: National Discount, Inc.

 Interval Estimate of the Population Mean: 

Unknown

Interval Estimate of  is:

$21,100 + $1,470
or $19,630 to $22,570

We are 95% confident that the interval contains

the
population mean.

© 2003 South-Western /Thomson Learning™

13
 Using Excel to Construct
a Confidence Interval: Large-Sample Case
 Formula Worksheet
A B C
1 Income Sample Size 36
2 25,600 Sample Mean =AVERAGE(A2:A37)
3 19,615 Sample Std. Deviation =STDEV(A2:A37)
4 20,035
5 21,735 Confidence Coefficient 0.95
6 17,600 Level of Signif. (alpha) =1-C5
7 26,080 z Value =NORMSINV(1-C5/2)
8 28,925
9 25,350 Standard Error =C3/SQRT(C1)
10 15,900 Margin of Error =C7*C9
11 17,550
12 14,745 Point Estimate =C2
13 20,000 Lower Limit =C12-C10
14 19,250 Upper Limit =C12+C10
Note: Rows 15-37 are not shown.
© 2003 South-Western /Thomson Learning™
14
 Using Excel to Construct
a Confidence Interval: Large-Sample Case
 Value Worksheet
A B C
1 Income Sample Size 36
2 25,600 Sample Mean 21,100
3 19,615 Sample Std. Deviation 4500
4 20,035
5 21,735 Confidence Coefficient 0.95
6 17,600 Level of Signif. (alpha) 0.05
7 26,080 z Value 1.960
8 28,925
9 25,350 Standard Error 749.992
10 15,900 Margin of Error 1469.955
11 17,550
12 14,745 Point Estimate 21,100.39
13 20,000 Lower Limit 19,630.43
14 19,250 Upper Limit 22,570.34
Note: Rows 15-37 are not shown.
© 2003 South-Western /Thomson Learning™
15
 Interval Estimation of a Population Mean:
Small-Sample Case (n < 30)
 Population is Not Normally Distributed
The only option is to increase the sample size to
n > 30 and use the large-sample interval-
estimation
procedures.
 Population is Normally Distributed and  is Assumed
Known
The large-sample interval-estimation procedure can
be used.
 Population is Normally Distributed and  is Estimated
by s
The appropriate interval estimate is based on a
probability distribution known as the t distribution.

© 2003 South-Western /Thomson Learning™

16
 t Distribution

 The t distribution is a family of similar

probability distributions.
 A specific t distribution depends on a
parameter known as the degrees of freedom.
 As the number of degrees of freedom
increases, the difference between the t
distribution and the standard normal
probability distribution becomes smaller and
smaller.
 A t distribution with more degrees of freedom
has less dispersion.
 The mean of the t distribution is zero.

© 2003 South-Western /Thomson Learning™

17
 Interval Estimation of a Population Mean:
Small-Sample Case (n < 30) with 
Estimated by s
 Interval Estimate
s
x t /2
n
where 1 - = the confidence coefficient
t/2 = the t value providing an area
of /2 in the upper tail of a t
distribution
with n - 1 degrees of freedom
s = the sample standard deviation

© 2003 South-Western /Thomson Learning™

18
 Example: Apartment Rents

 Interval Estimation of a Population Mean:

Small-Sample Case (n < 30) with  Estimated by
s
A reporter for a student newspaper is writing
an
article on the cost of off-campus housing. A
sample of 10 one-bedroom units within a half-mile
of campus resulted in a sample mean of $550 per
month and a sample standard deviation of $60.
Let us provide a 95% confidence interval
estimate of the mean rent per month for the
population of one-bedroom units within a half-mile
of campus. We’ll assume this population to be
normally distributed.
© 2003 South-Western /Thomson Learning™
19
 Example: Apartment Rents

 t Value
At 95% confidence, 1 -  = .95,  = .05, and /2
= .025.
t.025 is based on n - 1 = 10 - 1 = 9 degrees of
freedom.
Degrees Area in Upper Tail
InFreedom
of the t distribution
.10 .05table .025
we see that.01 t.025.005
=
2.262.. . . . . .
7 1.415 1.895 2.365 2.998 3.499
8 1.397 1.860 2.306 2.896 3.355
9 1.383 1.833 2.262 2.821 3.250
10 1.372 1.812 2.228 2.764 3.169
. . . . . .

© 2003 South-Western /Thomson Learning™

20
 Example: Apartment Rents

 Interval Estimation of a Population Mean:

Small-Sample Case (n < 30) with  Estimated
by s s
x t.025
n
60
550 2.262
10

550 + 42.92
or $507.08 to $592.92
We are 95% confident that the mean rent per
month for the population of one-bedroom units
within a half-mile of campus is between
$507.08 and $592.92.
© 2003 South-Western /Thomson Learning™
21
 Using Excel to Construct
a Confidence Interval:  Estimated by s
 Formula Worksheet
A B C
1 Rent Sample Size 10
2 600 Sample Mean =AVERAGE(A2:A11)
3 635 Sample Std. Deviation =STDEV(A2:A11)
4 550
5 535 Confidence Coefficient 0.95
6 465 Level of Signif. (alpha) =1-C5
7 625 Degrees of Freedom =C1-1
8 570 t Value =TINV(C6,C7)
9 465
10 550 Standard Error =C3/SQRT(C1)
11 505 Margin of Error =C8*C10
12
13 Point Estimate =C2
14 Lower Limit =C13-C11
15 Upper Limit =C13+C11

© 2003 South-Western /Thomson Learning™

22
 Using Excel to Construct
a Confidence Interval:  Estimated by s
 Value Worksheet
A B C
1 Rent Sample Size 10
2 600 Sample Mean 550.00
3 635 Sample Std. Deviation 60.05
4 550
5 535 Confidence Coefficient 0.95
6 465 Level of Signif. (alpha) 0.05
7 625 Degrees of Freedom 9
8 570 t Value 2.2622
9 465
10 550 Standard Error 18.988
11 505 Margin of Error 42.955
12
13 Point Estimate 550.00
14 Lower Limit 507.05
15 Upper Limit 592.95

© 2003 South-Western /Thomson Learning™

23
 Sample Size for an Interval Estimate
of a Population Mean
 Let E = the maximum sampling error
mentioned in the precision statement.
 E is the amount added to and subtracted from
the point estimate to obtain an interval
estimate.
 E is often referred to as the margin of error.

 We have E z /2
n

 Solving for n we have

( z / 2 ) 2  2
n
E2

© 2003 South-Western /Thomson Learning™

24
 Example: National Discount, Inc.

 Sample Size for an Interval Estimate of a

Population Mean
Suppose that National’s management
team wants an estimate of the population
mean such that there is a .95 probability that
the sampling error is $500 or less.
How large a sample size is needed to
meet the required precision?

© 2003 South-Western /Thomson Learning™

25
 Example: National Discount, Inc.

 Sample Size for Interval Estimate of a Population

Mean 
z /2 500
n

At 95% confidence, z.025 = 1.96.

Recall that = 4,500.
Solving for n we have2
(1.96) (4,500)2
n 2
311.17
(500)

We need to sample 312 to reach a desired

precision of
+ $500 at 95% confidence.
© 2003 South-Western /Thomson Learning™
26
 Interval Estimation
of a Population Proportion
 Interval Estimate

p (1  p )
p z / 2
n

where: 1 - is the confidence coefficient

z/2 is the z value providing an area
of
/2 in the upper tail of the
standard p normal probability
distribution
is the sample proportion

© 2003 South-Western /Thomson Learning™

27
 Example: Political Science, Inc.

 Interval Estimation of a Population Proportion

Political Science, Inc. (PSI) specializes in
voter polls and surveys designed to keep
political office seekers informed of their position
in a race. Using telephone surveys,
interviewers ask registered voters who they
would vote for if the election were held that
day.
In a recent election campaign, PSI found
that 220 registered voters, out of 500
contacted, favored a particular candidate. PSI
wants to develop a 95% confidence interval
estimate for the proportion of the population of
registered voters that favors the candidate.

© 2003 South-Western /Thomson Learning™

28
 Example: Political Science, Inc.

 Interval Estimate of a Population Proportion

p (1  p )
p z / 2
n
where: n = 500, p = 220/500 = .44, z/2 = 1.96

. 44 (1 . 44 )
. 44 1. 96
500
.44 + .0435
PSI is 95% confident that the proportion of all voters
that favors the candidate is between .3965
and .4835.

© 2003 South-Western /Thomson Learning™

29
 Using Excel to Construct a Confidence
Interval
 Formula Worksheet
A B C
1 Favored Sample Size 500
2 Yes Total Yes =COUNTIF(A2:A501,"Yes")
3 Yes Sample Proportion =C2/C1
4 No
5 Yes Confid. Coefficient 0.95
6 No Lev. of Signif. (alpha) =1-C5
7 No z Value =NORMSINV(1-C6/2)
8 No
9 No Standard Error =SQRT(C3*(1-C3)/C1)
10 Yes Margin of Error =C7*C9
11 No
12 Yes Point Estimate =C3
13 No Lower Limit =C12-C10
14 No Upper Limit =C12+C10

Note: Rows 15-501 are not shown.

© 2003 South-Western /Thomson Learning™
30
 Using Excel to Construct a Confidence
Interval
 Value Worksheet
A B C
1 Favored Sample Size 500
2 Yes Total Yes 220
3 Yes Sample Proportion 0.4400
4 No
5 Yes Confid. Coefficient 0.95
6 No Lev. of Signif. (alpha) 0.05
7 No z Value 1.9600
8 No
9 No Standard Error 0.02220
10 Yes Margin of Error 0.04351
11 No
12 Yes Point Estimate 0.4400
13 No Lower Limit 0.3965
14 No Upper Limit 0.4835

Note: Rows 15-501 are not shown.

© 2003 South-Western /Thomson Learning™
31
 Sample Size for an Interval Estimate
of a Population Proportion
 Let E = the maximum sampling error
mentioned in the precision statement.
 We have
p (1  p )
E z / 2
n
 Solving for n we have
( z / 2 ) 2 p (1  p )
n
E2

© 2003 South-Western /Thomson Learning™

32
 Example: Political Science, Inc.

 Sample Size for an Interval Estimate of a

Population Proportion
Suppose that PSI would like a .99
probability that the sample proportion is within
+ .03 of the population proportion.
How large a sample size is needed to
meet the required precision?

© 2003 South-Western /Thomson Learning™

33
 Example: Political Science, Inc.

 Sample Size for Interval Estimate of a Population

Proportion
At 99% confidence, z.005 = 2.576.

( z / 2 ) 2 p (1  p ) ( 2. 576) 2 (. 44)(. 56)

n 2
 2
1817
E (. 03)
Note: We used .44 as the best estimate of p in the
above expression. If no information is available
about p, then .5 is often assumed because it
provides
the highest possible sample size. If we had used
p = .5, the recommended n would have been
1843.

© 2003 South-Western /Thomson Learning™

34
 End of Chapter 8

© 2003 South-Western /Thomson Learning™

35