Chapter 9 – Estimating the

Value of a Parameter
OUTLINE The shape of the distribution of the sample mean
9.1 Estimating a Population Proportion is approximately normal if the population from
9.2 Estimating a Population Mean which the sample was drawn if approximately
9.3 Putting It Together: Which Procedure normal or if the sample size is large. We learned
Do I Use? in Section 8.2 that p̂ is also a random variable
9.4 Estimating a Population Mean whose mean is p and whose standard deviation is
p 1  p 
. If np 1  p   10 , the distribution
n
Putting It Together
Chapters 1 through 7 laid the groundwork of the random variable p̂ is approximately
for the remainder of the course. These normal. For both the distribution of the sample
chapters dealt with data collection (Chapter mean and sample proportion, the sample size
1), descriptive statistics (Chapters 2 must be small relative to the size of the
through 4), and probability (Chapters 5 population  n  0.05 N  . This requirement is
through 7). Chapter 8 formed a bridge necessary so that the sampled values are
between probability and statistical independent of each other.
inference by giving us models that can be We now discuss inferential statistics—the
used to make probability statements about process of generalizing information obtained
the sample mean and sample proportion. from a sample to a population. We will study two
areas of inferential statistics:
We know from Section 8.1 that the sample
mean is a random variable and has a Estimation—sample data are used to estimate the
distribution associated with it. This value of unknown parameters such as  or p.
distribution is called the sampling
distribution of the sample mean. The mean
Hypothesis testing—statements regarding a
of this distribution is equal to the mean of
characteristic of one or more populations are
the population from which the sample was
tested using sample data.

drawn, , and the standard deviation is
n In this chapter, we discuss estimation of an
unknown parameter, and in the next chapter we
discuss hypothesis testing.

Copyright © 2024 Pearson Education, Inc.

193
Chapter 9: Estimating the Value of a Parameter

Section 9.1
Estimating a Population Proportion
Objectives
 Obtain a Point Estimate for the Population Proportion

 Construct and Interpret a Confidence Interval for the Population Proportion

 Determine the Sample Size Necessary for Estimating a Population Proportion

within a Specified Margin of Error

Objective 1: Obtain a Point Estimate for the Population Proportion

1) State the definition of a point estimate.

Example 1 Obtaining a Point Estimate of a Population Proportion

The Gallup Organization conducted a poll in which a simple random sample of 1015 Americans aged 18
and older were asked, “Do you consider the amount of income tax that you have to pay as too high?” Of
the 1015 adult Americans surveyed, 511 said yes. Obtain a point estimate for the proportion of Americans
aged 18 and older who believe that the amount of income tax they pay is too high.

Objective 2: Construct and Interpret a Confidence Interval for the Population Proportion

Statistics such as p̂ vary from sample to sample.

Due to variability in the sample proportion, we report a range (or interval) of values, including a measure
of the likelihood that the interval includes the unknown population parameter.

Copyright © 2024 Pearson Education, Inc.

194
 Section 9.1: Estimating a Population Proportion

Watch the video, which discusses the logic behind the construction of confidence intervals for a
population proportion. Answer the following as you watch.
2) Give the definition for a confidence interval for an unknown parameter.

3) What does the level of confidence represent?

4) What is the form of confidence interval estimates for a population parameter?

Note: Review of the Sampling Distribution of the Sample Proportion

 The shape of the distribution of all possible sample proportions is approximately normal provided
np 1  p   10 and the sample size is no more than 5% of the population size. That is, n  0.05 N
.

 The mean of the distribution of the sample proportions equals the population proportion. That is,
 p̂  p .

 The standard deviation of the distribution of the sample proportion (the standard error) is

 p̂ 
n

5) How is the margin of error computed for a 95% confidence interval for a population proportion?

Copyright © 2024 Pearson Education, Inc.

195
Chapter 9: Estimating the Value of a Parameter

 For a 95% confidence interval, any sample proportion that lies within 1.96 standard errors of the
population proportion will result in a confidence interval that includes p. This will happen in 95%
of all possible samples.

 Any sample proportion that is more than 1.96 standard errors from the population proportion will
result in a confidence interval that does not contain p. This will happen in 5% of all possible
samples (those sample proportions in the tails of the distribution).

 Whether a confidence interval contains the population parameter depends solely on the value of
the sample statistic.

 Any sample statistic that is in the tails of the sampling distribution will result in a confidence
interval that does not include the population parameter.

Key Ideas Regarding Confidence Intervals

 A confidence interval for an unknown parameter consists of an interval of numbers based on a
point estimate.

 The level of confidence represents the expected proportion of intervals that will contain the
parameter if a large number of different samples is obtained. The level of confidence is denoted
(1−)⋅100%.

 Whether a confidence interval contains the population parameter depends solely on the value of
the sample statistic. Any sample statistic that is in the tails of the sampling distribution will result
in a confidence interval that does not include the population parameter.

Answer the following after Activity 1: Illustrating the Meaning of Level of Confidence Using
Simulation
6) What proportion of the 95% confidence intervals generated by simulation would you expect to contain
the population parameter p? What proportion of the 99% confidence intervals would you expect to
contain the population parameter p?

Watch the video to reinforce the concepts learned from Activity 1.

Copyright © 2024 Pearson Education, Inc.

196
 Section 9.1: Estimating a Population Proportion

Answer the following after watching the video on reinforcing the meaning of the level of confidence.
7) What does the “95%” in a 95% confidence interval represent?

1    100% of all sample proportions will result in confidence intervals that contain the population
proportion. The sample proportions that are in the tails of the distribution, will result in confidence
intervals that contain the population proportion.

8) What does the critical value of a distribution represent?

9) As the level of confidence increases, what happens to the critical value?

10) List the critical value associated with the given level of confidence.
A) 90% B) 95% C) 99%

11) State the interpretation of a confidence interval.

Copyright © 2024 Pearson Education, Inc.

197
Chapter 9: Estimating the Value of a Parameter

Example 2 Interpreting a Confidence Interval

The Gallup Organization conducted a poll in which a simple random sample of 1015 Americans aged 18
and older were asked, “Do you consider the amount of income tax that you have to pay as too high?” Of
the 1015 adult Americans surveyed, 511 said yes. We learned from Example 1 that the proportion of
those surveyed who responded yes was 0.503. Gallup reported its “survey methodology” as follows:

Results are based on telephone interviews with a random sample of 1015 national adults, aged 18 and
older. For results based on the total sample of national adults, one can say with 95% confidence that the
maximum margin of sampling error is 3.1 percentage points.

Determine and interpret the confidence interval for the proportion of Americans aged 18 and older who
believe the amount of federal income tax they have to pay is too high.

Note:
A 90% level of confidence does not tell us that there is a 90% probability that the parameter lies between
the lower and upper bound.
It means that the interval includes the unknown parameter for 90% of all samples.

Copyright © 2024 Pearson Education, Inc.

198
 Section 9.1: Estimating a Population Proportion

12) Explain the method for constructing a confidence interval about the population proportion, p.

Note: It must be the case that npˆ 1  pˆ   10 and n  0.05 N to construct this interval. Use p̂ in place of
p in the standard deviation. This is because p is unknown, and p̂ is the best point estimate of p.

Example 3 Constructing a Confidence Interval for a Population Proportion

In the Parent–Teen Cell Phone Survey conducted by Princeton Survey Research Associates International,
800 randomly sampled 16- to 17-year-olds living in the United States were asked whether they have ever
used their cell phone to text while driving. Of the 800 teenagers surveyed, 272 indicated that they text
while driving. Obtain a 95% confidence interval for the proportion of 16- to 17-year-olds who text while
driving.

Copyright © 2024 Pearson Education, Inc.

199
Chapter 9: Estimating the Value of a Parameter

13) State the formula for the margin of error, E, for a 1     100% confidence interval for a population
proportion.

Example 4 The Role of the Level of Confidence in the Margin of Error

In the Parent–Teen Cell Phone Survey conducted by Princeton Survey Research Associates International,
800 randomly sampled 16- to 17-year-olds living in the United States were asked whether they have ever
used their cell phone to text while driving. Of the 800 teenagers surveyed, 272 indicated that they text
while driving. From the last example, we concluded that we are 95% confident that the proportion of 16-
to 17-year olds who text while driving is between 0.307 and 0.373. Determine the effect on the margin of
error by increasing the level of confidence from 95% to 99%.

14) As the sample size, n, increases, what happens to the margin of error?

Copyright © 2024 Pearson Education, Inc.

200
 Section 9.1: Estimating a Population Proportion

Answer the following after doing Activity 2.

15) If the normality condition is not satisfied, how does the proportion of intervals that capture the
parameter compare to the level of confidence?

Objective 3: Determine the Sample Size Necessary for Estimating a Population Proportion
within a Specified Margin of Error

Watch the video for an explanation of where the formulas for determining sample size for estimating
a population proportion within a specified margin of error come from.

16) List the formula for the sample size required to obtain a 1     100% confidence interval for p with
a margin of error E, if p̂ is a prior estimate of p.

17) List the formula for the sample size required to obtain a 1     100% confidence interval for p with
a margin of error E, if a prior estimate of p is unavailable.

Copyright © 2024 Pearson Education, Inc.

201
Chapter 9: Estimating the Value of a Parameter

Example 5 Determining Sample Size

An economist wants to know if the proportion of the U.S. population who commutes to work via car-
pooling is on the rise. What size sample should be obtained if the economist wants an estimate within 2
percentage points of the true proportion with 90% confidence?

A) Assume that the economist uses the estimate of 10% obtained from the American Community Survey.

B) Assume that the economist does not use any prior estimates.

Copyright © 2024 Pearson Education, Inc.

202
Chapter 9: Estimating the Value of a Parameter

Section 9.2
Estimating a Population Mean
Objectives
 Obtain a Point Estimate for the Population Mean

 State Properties of Student’s t-Distribution

 Determine t-Values

 Construct and Interpret a Confidence Interval for a Population Mean

 Determine the Sample Size Necessary to Estimate a Population Mean Within a

Specified Margin of Error

Objective 1: Obtain a Point Estimate for the Population Mean

1) What is the point estimate for a population mean  ?

Example 1 Computing a Point Estimate of the Population Mean

The website fueleconomy.gov allows drivers to report the miles per gallon of their vehicle. The data in
Table 2 show the reported miles per gallon of 2019 Honda Accord Hybrid automobiles for 16 different
owners. Obtain a point estimate of the population mean miles per gallon of a 2019 Honda Accord Hybrid.

Table 2
42.3 43.6 48.2 40.0
39.0 42.9 44.8 43.0
40.7 38.6 39.0 44.5
35.5 45.1 35.4 42.2

Copyright © 2024 Pearson Education, Inc.

203
Chapter 9: Estimating the Value of a Parameter

Objective 2: State Properties of Student’s t-Distribution

Watch the video that discusses the logic behind constructing a confidence interval for the population
mean.

Recall that the distribution of x is approximately normal if the population from which the sample is
drawn is normal or the sample size is sufficiently large. In addition, the distribution of x has the same
mean as the parent population,  x   , and a standard deviation equal to the parent population’s standard

deviation divided by the square root of the sample size,  x  .
n


Using x  z  for a confidence interval for the mean presents a problem because it is not likely that
2 n
we know the population standard deviation   but not know the population mean    .

Using s as an estimate for  also presents a problem because the sample standard deviation, s, is a
statistic and therefore will vary from sample to sample. Using the normal model to determine the critical
value, z , in the margin of error does not take into account the additional variability introduced by using
2

s in place of  . A new model must be used to determine the margin of error in a confidence interval that
accounts for the additional variability. This leads to the story of William Gosset.

2) What was the name of the brewery that Gosset worked for? What pseudonym did he choose to publish
his results about a model that accounts for the additional variability introduced by using s in place of 
when determining margin of error.

3) Suppose that a simple random sample of size n is taken from a population. If the population from
x 
which the sample is drawn follows a normal distribution, what does the distribution of t  follow?
s
n

Copyright © 2024 Pearson Education, Inc.

204
 Section 9.2: Estimating a Population Mean

Watch the video that uses simulation to illustrate some of the work Gosset did by hand to develop
his sampling distribution – Student’s t-distribution..

4) State six properties of the t-distribution.

Objective 3: Determine t-Values

5) What does t represent?

Copyright © 2024 Pearson Education, Inc.

205
Chapter 9: Estimating the Value of a Parameter

Example 2 Finding t-Values

Find the t-value such that the area under the t-distribution to the right of the t-value is 0.10, assuming 15
degrees of freedom (df). That is, find t0.10 with 15 degrees of freedom.

If the degrees of freedom we desire are not listed in Table VII, choose the closest number in the “df”
column.

In addition, the last row of Table VII lists the z-values from the standard normal distribution. Use these
values when the degrees of freedom are more than 1000 because the t-distribution starts to behave like the
standard normal distribution as n increases.

Objective 4: Construct and Interpret a Confidence Interval for a Population Mean

6) List the three conditions required for constructing a confidence interval for a population mean  .

7) List the formulas for the lower bound and upper bound for a 1     100% confidence interval for the
population mean,  .

Copyright © 2024 Pearson Education, Inc.

206
 Section 9.2: Estimating a Population Mean

8) What does it mean when we say that the procedure for constructing a confidence interval is robust?

9) Because the sample mean and sample standard deviation are not resistant to outliers, sample data
should always be inspected for serious departures from normality and for outliers. What tools can be used
to check for serious departures from normality and for outliers?

Example 3 Constructing a Confidence Interval about a Population Mean

The website fueleconomy.gov allows drivers to report the miles per gallon of their vehicle. The data in
Table 2 show the reported miles per gallon of 2019 Honda Accord Hybrid automobiles for 16 different
owners. Treat the sample as a simple random sample of all 2019 Honda Acord Hybrid automobiles.
Construct a 95% confidence interval for the mean miles per gallon of a 2019 Honda Accord Hybrid.
Interpret the interval.

Table 2
42.3 43.6 48.2 40.0
39.0 42.9 44.8 43.0
40.7 38.6 39.0 44.5
35.5 45.1 35.4 42.2

The t-distribution gives a larger critical value than the z-distribution, so the width of the confidence
interval is wider when it is constructed using Student’s t-distribution.

Copyright © 2024 Pearson Education, Inc.

207
Chapter 9: Estimating the Value of a Parameter

Answer the following after Activity 1: When Model Requirements Fail.

10) What happens to the proportion of intervals that capture the population mean as the sample size
increases?

If the requirements to compute a t-interval are not met, one option is to use resampling methods, such as
bootstrapping. Bootstrapping is presented later in this chapter.

Objective 5: Determine the Sample Size Necessary to Estimate a Population Mean Within a
Specified Margin of Error

11) State the formula for margin of error when constructing a confidence interval about the population
mean.

12) List the formula for the sample size required to obtain a 1     100% confidence interval for  with
a margin of error E.

Example 4 Constructing a Confidence Interval about a Population Mean

We again consider the problem of estimating the miles per gallon of a 2019 Honda Accord Hybrid. How
large a sample is required to estimate the mean miles per gallon within 0.5 miles per gallon with 95%
confidence? Note: The sample standard deviation is s=3.51 mpg.

Copyright © 2024 Pearson Education, Inc.

208