Learning Objectives

LO9-1 Compute and interpret a point estimate of

a population mean
LO9-2 Compute and interpret a confidence
interval for a population mean
LO9-3 Compute and interpret a confidence
interval for a population proportion
LO9-4 Calculate the required sample size to
Estimation and Confidence Intervals estimate a population proportion or
population mean
Chapter 9
LO9-5 Adjust a confidence interval for finite
populations

9-1 Copyright © 2018 by McGraw-Hill Education. All rights reserved. 9-2 Copyright © 2018 by McGraw-Hill Education. All rights reserved.

Point Estimate Confidence Intervals

 A point estimate is a single value (statistic) used to estimate a  A confidence interval is a range of values within which the
population value (parameter) population parameter is expected to occur

POINT ESTIMATE The statistic, computed from sample information, CONFIDENCE INTERVAL A range of values constructed from sample
that estimates a population parameter. data so that the population parameter is likely to occur within that
range at a specified probability. The specified probability is called the
level of confidence.
 Example
 The factors that determine the width of a confidence interval
 Suppose the Bureau of Tourism for Barbados wants an for a mean are
estimate of the mean amount spent by tourists visiting that  The number of observations in the sample, n
country. They randomly select 500 tourists as they depart and
 The variability in the population, usually estimated by the
ask these tourists about their spending while there. The mean sample standard deviation, s
amount spent by the sample of 500 tourists serves as an
 The desired level of confidence
estimate of the unknown population parameter.

9-3 Copyright © 2018 by McGraw-Hill Education. All rights reserved. 9-4 Copyright © 2018 by McGraw-Hill Education. All rights reserved.
Level of Confidence, Known Finding a Value of z
 To determine the confidence limits when the population  The method for finding z for a 95% confidence interval is
standard deviation is known, we use the z distribution  Divide the confidence interval in half, .9500/2 = .4750
 The formula is  Find the value .4750 in the body of the table
 Identify the row and column and add the values
 The probability of finding a value between 0 and 1.96 is
.4750
x  sample mean
 So the probability of finding a value between +/- 1.96 is
z  z - value for a particular confidence level
.9500
σ  the population standard deviation
n  the number of observations in the sample

9-5 Copyright © 2018 by McGraw-Hill Education. All rights reserved. 9-6 Copyright © 2018 by McGraw-Hill Education. All rights reserved.

Level of Confidence, z Example Level of Confidence, Unknown

 To determine the confidence limits when the population
The American Management Association is studying the income of store standard deviation is unknown, we use the t distribution
managers in the retail industry. A random sample of 49 managers reveals a
sample mean of $45,420. The standard deviation of the population is $2,050.  The formula is
1. What is the population mean?
2. What is a reasonable range of values for the population mean?
3. How do we interpret these results?

 Example
1. We do not know the population mean, so we can use the sample mean,
$45,420 as our best estimate.
 The Dean of the Business College wants to estimate the mean
number of hours full-time students work at paying jobs each
2. The AMA decides to use a 95% level of confidence, so use equation (9-1), week. He randomly selects a sample of 30 students and asks
σ $2.050
x z
n
= $45,420 1.96
49
= $45,420 $574 them how many hours they worked last week. He can
calculate the sample mean, but it is unlikely he would know the
3. The confidence interval is from $44,846 and $45,994, the value $574 is
called the margin of error.
population standard deviation required for formula 9-1.

9-7 Copyright © 2018 by McGraw-Hill Education. All rights reserved. 9-8 Copyright © 2018 by McGraw-Hill Education. All rights reserved.
Characteristics of the t Distribution Finding a Value of t
 The t distribution is a continuous distribution  First assume the population is normal
 It is mound-shaped and symmetrical  Using Appendix B.5, move across the columns identified
 It is flatter, or more spread out, than the standard normal for confidence intervals
distribution  In the next example, we want to use the 95% level of
 There is a family of t distributions, depending on the confidence, so move to that column
number of degrees of freedom  Then find df, the degrees of freedom, n-1

9-9 Copyright © 2018 by McGraw-Hill Education. All rights reserved. 9-10 Copyright © 2018 by McGraw-Hill Education. All rights reserved.

Level of Confidence, t Example Confidence Intervals for Proportions

PROPORTION The fraction, ratio, or percent indicating the part of the
A tire manufacturer wishes to investigate the tread life of its tires. A sample
sample or the population having a particular trait of interest.
of 10 tires driven 50,000 miles revealed a sample mean of 0.32 inch of tread
remaining with a standard deviation 0.09 inch. Construct a 95% confidence  A sample proportion, p, is found by x, the number of
interval for the population mean.
successes, divided by n, the number of observations
Would it be reasonable for the manufacturer to conclude that after 50,000
miles the population mean amount of tread remaining is 0.30 inch?

x t
s
= 0.32 2.262
0.09
= 0.32 0.064
 Examples
n 10  Southern Tech career services reports that 80% of its
The endpoints of the confidence interval are 0.256 and 0.384. The margin of graduates enter the job market in a position related to their
error is 0.064. The manufacturer can be reasonably sure (95% confident) that field of study
the mean remaining tread depth is between 0.256 and 0.384 inch. Because  A recent study of married men between the ages 35 and 50
the value 0.30 is in this interval, it is possible that the mean of the population
is 0.30.
found that 63% felt that both partners should earn a living

9-11 Copyright © 2018 by McGraw-Hill Education. All rights reserved. 9-12 Copyright © 2018 by McGraw-Hill Education. All rights reserved.
Confidence Intervals for Proportions Confidence Interval,
 A population proportion is identified by The union representing the Bottle Blowers of America (BBA) is considering
a proposal to merge with the Teamsters Union. At least three-fourths of the
 Two requirements BBA membership must approve any merger. A random sample of 2,000
 The binomial conditions have been met current members reveals 1,600 plan to vote for the merger proposal. What
is the estimate of the population proportion? Can you conclude that the
 The values n and n(1- ) should both be greater than necessary proportion of BBA members favor the merger? Why?
or equal to 5
x ,
First, calculate the sample proportion, p = = = .80
 We construct a confidence interval for a population n ,

proportion with the following formula Next, use formula (9-4) to determine the 95% confidence interval,

p 1 p .80 1 .80
p z = .80 1.96 = .80 .018
n 2,000
The endpoints of the confidence interval are .782 and .818, so we conclude
the merger will likely pass because the interval estimate includes values
greater than 75% of the union membership.
9-13 Copyright © 2018 by McGraw-Hill Education. All rights reserved. 9-14 Copyright © 2018 by McGraw-Hill Education. All rights reserved.

Sample Size to Estimate a Population Mean

Determining Sample Size for Means Example
 There are three factors that determine the sample size when
we wish to estimate the mean A student in public administration wants to estimate the mean monthly
earnings of city council members in large cities. She can tolerate a margin of
 The margin of error, E error of $100 in estimating the mean. She would also prefer to report the
 The desired level of confidence, for example 95% interval estimate with a 95% level of confidence. The student found a report
by the Department of Labor that reported a standard deviation of $1,000.
 The variation in the population What is the required sample size?
 The formula to determine the sample size for the mean is

. $ ,
n= 2 2 19.6 2 384.16
$

The computed value of 384.16 is rounded up to 385. A sample size of 385 is

 The result is not always a whole number; the usual practice is required to meet the specifications.

to round up any fractional result to the next whole number.

9-15 Copyright © 2018 by McGraw-Hill Education. All rights reserved. 9-16 Copyright © 2018 by McGraw-Hill Education. All rights reserved.
 Sample Size for the Population Proportion
Determining Sample Size for Proportions Example
 There are three factors that determine the sample size
when we wish to estimate a proportion The student in the previous example also wants to estimate the proportion
of cities that have private refuse collectors. The student wants to estimate
 The margin of error, E the population proportion with a margin of error of .10, prefers a level of
confidence of 90%, and has no estimate for the population proportion.
 The desired level of confidence What is the required sample size?
 A value for to calculate the variation in the population
 The formula to determine the sample size for a
proportion is 1.645
n = (.50)(1-.50) 2 = 67.65
.10

The student needs a random sample of 68 cities.

9-17 Copyright © 2018 by McGraw-Hill Education. All rights reserved. 9-18 Copyright © 2018 by McGraw-Hill Education. All rights reserved.

Finite Population Adjustment Finite Population Adjustment Example

 A population that has a fixed upper bound is finite There are 250 families residing in Scandia, Pennsylvania. A random sample of 40
of these families revealed the mean annual church contribution was $450 and
 For a finite population, the standard error is adjusted by the standard deviation of this was $75.
1. What is the population mean? What is the best estimate of the population
the finite-population correction factor, mean?
2. Develop a 90% confidence interval for the population mean.
3. Using the confidence interval, explain why the population mean could be
 This will make your estimate more precise by reducing $445. Could the population mean be $425? Why?
the standard error and resulting in a smaller range of
values in estimating the population mean 1. We do not know the population mean. The best estimate is $450.
2. The formula is
s N n $75 250 40
x t = $450 1.685 = $450 $18.35
n N 1 40 250 1
3. The endpoints are $431.65 and $468.35, so the population mean could be
$445. It is not likely the population mean is $425 since $425 is not within
the confidence interval.

9-19 Copyright © 2018 by McGraw-Hill Education. All rights reserved. 9-20 Copyright © 2018 by McGraw-Hill Education. All rights reserved.