Interval Estimation

Dr. Bui Thanh Tu

Email: tubt@hus.edu.vn
Introduction
Recall:
The sample mean 𝑥᪄ is a point estimator of the population mean 𝜇
The sample proportion 𝑝᪄ is a point estimator of the population proportion 𝑝.
Point estimator cannot provide the exact value of the population parameter

In this Chapter:
An interval estimate is computed by:
Point estimate ± Margin of error
How to compute margin of error of a population mean 𝜇 and a population proportion 𝑝.
The general form of an interval estimate of a population mean is
𝑥᪄ ± Margin of error
The general form of an interval estimate of a population proportion is
𝑝᪄ ± Margin of error
Margin of Error and the Interval Estimate
Population of data is normally
distributed with a standard
deviation of 𝜎 = 20

Sample size 𝑛 = 100;

Standard error:
𝜎𝑥᪄ = 𝜎/ 𝑛 = 20/ 100 = 2

95% of the values of any

normally distributed random
variable are between ±1.96 standard
deviations of the mean.

That means 95% of

the sample mean ᪄𝑥 values must be
within ±1.96 𝜎𝑥᪄
Interval contains the population mean

From the Figure:

Interval formed by 𝑥᪄1 and 𝑥᪄2 include the
population mean 𝜇.
Interval formed by 𝑥᪄3 does not include 𝜇.

Any sample mean 𝑥᪄ that is within the

darkly shaded region of Figure 8.3 will
provide an interval that contains the
population mean 𝜇

We say that this interval has been established

at the 95% confidence level.

The value .95 is referred to as the confidence

coefficient

The interval 𝑥᪄ ± margin of error is called the

95% confidence interval
INTERVAL ESTIMATE OF A POPULATION MEAN
INTERVAL ESTIMATE OF A POPULATION MEAN: 𝜎 How to construct a 95% confidence interval?
KNOWN
𝜎 The confidence coefficient is (1 − 𝛼) = .95
𝑥᪄ ± 𝑧𝛼/2
𝑛 Thus, 𝛼 = .05.

where (1 − 𝛼) is the confidence coefficient Using the standard normal probability table or
calculator, an area of 𝛼/2 = .05/2 = .025 in the
𝑧𝛼/2 is the 𝑧 value providing an area of 𝛼/2 in the upper upper tail
tail of the standard normal probability distribution
This provides 𝑧.025 = 1.96.
Margin of error
𝜎 Thus, the margin of error is 3.92
𝑧𝛼/2
𝑛
Other confidence levels such as 90% and 99% may be considered.
Note and comments
The interval estimation procedure is based on the With n in the denominator in
assumption that the population standard deviation s is 𝜎
𝑥᪄ ± 𝑧𝛼/2
known from historical data 𝑛
So technically we don’t mean that s is actually known
a larger sample size will provide a smaller margin of error,
with certainty. We just mean that we obtained a good
a narrower interval, and greater precision.
estimate of the standard deviation prior to sampling and
thus we won’t be using the same sample to estimate
The procedure for determining the size of a simple
both the population mean and the population standard
random sample necessary to obtain a desired precision
deviation
is discussed later

In most applications, a sample size of n ≥ 30 is adequate

to develop an interval estimate of a population mean.

If the population is not normally distributed but is

roughly symmetric, sample sizes as small as 15 can be
expected to provide good approximate confidence
intervals.
Exercise: calculate confident interval
1. A simple random sample of 50 items from a population Margin of error
𝜎
with 𝜎 = 6 resulted in a sample mean of 32. 𝑧𝛼/2
a. Provide a 90% confidence interval for the population 𝑛
mean.
b. Provide a 95% confidence interval for the population
mean.
c. Provide a 99% confidence interval for the population
mean.
Exercise: margin of error
Studies show that massage therapy has a variety of
health benefits and it is not too expensive (The Wall
Street Journal, March 13, 2012). A sample of 10 typical
one-hour massage therapy sessions showed an average
charge of $59. The population standard deviation for a
one-hour session is 𝜎 = $5.50.
a. What assumptions about the population should we be
willing to make if a margin of error is desired?
b. Using 95% confidence, what is the margin of error?
c. Using 99% confidence, what is the margin of error?
Homework: calculate margin of error
P1. A simple random sample of 40 items resulted in
a sample mean of 25 . The population standard
deviation is 𝜎 = 5.
a. What is the standard error of the mean, 𝜎𝑥᪄ ?
b. At 95% confidence, what is the margin of error?
Homework
10. Costs are rising for all kinds of medical care. The mean monthly rent at assisted-living facilities was reported
to have increased 17% over the last five years to $3486 (The Wall Street Journal, October 27, 2012). Assume this
cost estimate is based on a sample of 120 facilities and, from past studies, it can be assumed that the population
standard deviation is 𝜎 = $650.
a. Develop a 90% confidence interval estimate of the population mean monthly rent.
b. Develop a 95% confidence interval estimate of the population mean monthly rent.
c. Develop a 99% confidence interval estimate of the population mean monthly rent.
d. What happens to the width of the confidence interval as the confidence level is increased? Does this seem
reasonable? Explain.
Population Mean: σ Unknown
Purpose: developing an interval estimate of a
population mean, in the case that do not have a
good estimate of population.
We need to estimate both 𝜇 and 𝜎. This situation
represents the 𝜎 unknown case.
When 𝑠 is used to estimate 𝜎, the margin of error
and the interval estimate for the population mean
are based on a probability distribution known as
the 𝑡 distribution
the 𝑡 distribution
The difference between the t distribution and the
standard normal distribution becomes smaller
and smaller as the number of degrees of freedom
increases

We place a subscript on 𝑡 to indicate the area

in the upper tail of the 𝑡 distribution

𝑧.025 to indicate the 𝑧 value providing a .025

area in the upper tail of a standard normal
distribution

𝑡𝛼/2 to represent a
𝑡 value with an
area of 𝛼/2 in the
upper tail of the 𝑡
distribution
the 𝑡 distribution
Each row in the table corresponds to a
separate 𝑡 distribution with the degrees
of freedom shown. For example, for a 𝑡
distribution with 9 degrees of freedom,
𝑡.025 = 2.262. Similarly, for a 𝑡
distribution with 60 degrees of freedom,
𝑡.025 = 2.000. As the degrees of freedom
continue to increase, 𝑡.025 approaches
𝑧.025 = 1.96. In fact, the standard
normal distribution 𝑧 values can be
found in the infinite degrees of freedom
row (labeled ∞ ) of the 𝑡 distribution
table.
Margin of Error and the Interval Estimate
An interval estimate of a population mean for the 𝜎
known case is
𝜎
𝑥᪄ ± 𝑧𝛼/2
𝑛

To compute an interval estimate of 𝜇 for the 𝜎

unknown case, the sample standard deviation 𝑠 is
used to estimate 𝜎, and 𝑧𝛼/2 is replaced by the 𝑡
distribution value 𝑡𝛼/2 .

INTERVAL ESTIMATE OF A POPULATION

MEAN: 𝜎 UNKNOWN
𝑠
𝑥᪄ ± 𝑡𝛼/2
𝑛
where 𝑠 is the sample standard deviation,
(1 − 𝛼) is the confidence coefficient,
𝑡𝛼/2 is the 𝑡 value providing an area of 𝛼/2 in the
upper tail of the 𝑡 distribution with 𝑛 − 1 degrees of
freedom.
Degrees of freedom
The reason the number of degrees of freedom
associated with the 𝑡 value is 𝑛 − 1 concerns the use
of 𝑠 as an estimate of the population standard
deviation 𝜎. The expression for the sample
standard deviation is

σ 𝑥𝑖 − 𝑥᪄ 2
𝑠=
𝑛−1

Degrees of freedom refer to the number of

independent pieces of information that go into the
computation of σ 𝑥𝑖 − 𝑥᪄ 2 .
Example
we will consider a study
designed to estimate the mean
credit card debt for the
population of U.S.
households.
A sample of 𝑛 = 70
households provided the
credit card balances shown in
Table. For this situation, no
previous estimate of the
population standard deviation
𝜎 is available
Example
Sample mean
𝑥᪄ = $9312
Sample standard deviation
𝑠 = $4007
With 95% confidence and
𝑛 − 1 = 69 degrees of freedom
Use the table, the value shown is 𝑡.025 = 1.995.

We use
𝑠 4007
𝑥᪄ ± 𝑡𝛼/2 9312 ± 1.995
𝑛 70
9312 ± 955

The point estimate of the population mean is

$9312, the margin of error is $955
The 95% confidence interval is
9312 − 955 = $8357 to 9312 + 955 = $10,267.
Practical Advice
In most applications, a sample size of n ≥ 30 is adequate
when using expression (8.2)
to develop an interval estimate of a population mean.
However, if the population distribution is highly skewed
or contains outliers, most statisticians would recommend
increasing
the sample size to 50 or more. If the population is not
normally distributed but is roughly
symmetric, sample sizes as small as 15 can be expected
to provide good approximate
confidence intervals.
Summary of Interval Estimation Procedures
Exercise
Find the 𝑡 value(s) for each of the following cases.
a. Upper tail area of .025 with 12 degrees of
freedom
b. Lower tail area of .05 with 50 degrees of freedom
c. Upper tail area of .01 with 30 degrees of freedom
d. Where 90% of the area falls between these two 𝑡
values with 25 degrees of freedom
e. Where 95% of the area falls between these two 𝑡
values with 45 degrees of freedom
Homework
For a 𝑡 distribution with 16 degrees of freedom, find
the area, or probability, in each region.
a. To the right of 2.120
b. To the left of 1.337
c. To the left of -1.746
d. To the right of 2.583
e. Between -2.120 and 2.120
f. Between -1.746 and 1.746
Homework
The following sample data are from a normal
population: 10,8,12,15,13,11,6,5.
a. What is the point estimate of the population mean?
b. What is the point estimate of the population
standard deviation?
c. With 95% confidence, what is the margin of error
for the estimation of the population mean?
d. What is the 95% confidence interval for the
population mean?
Homework
Sales personnel for Skillings Distributors submit
weekly reports listing the customer contacts made
during the week. A sample of 65 weekly reports
showed a sample mean of 19.5 customer contacts per
week. The sample standard deviation was 5.2.
Provide 90% and 95% confidence intervals for the
population mean number of weekly customer
contacts for
the sales personnel
Determining the Sample Size
Let 𝐸 = the desired margin of error:
𝜎
𝐸 = 𝑧𝛼/2
𝑛
Solving for 𝑛, we have
𝑧𝛼/2 𝜎
𝑛=
𝐸
Squaring both sides, sample size provides the desired
margin of error at the chosen confidence level.

SAMPLE SIZE FOR AN INTERVAL ESTIMATE OF

POPULATION MEAN
2
𝑧𝛼/2 𝜎 2
𝑛=
𝐸2
Exercise
The range for a set of data is estimated to be 36 .
a. What is the planning value for the population
standard deviation?
b. At 95% confidence, how large a sample would
provide a margin of error of 3 ?
c. At 95% confidence, how large a sample would
provide a margin of error of 2 ?
Homework
How large a sample should be selected to provide a
95% confidence interval with a margin of error of
10 ? Assume that the population standard deviation
is 40 .
Homework
The U.S. Energy Information Administration (US EIA)
reported that the average price for a gallon of regular
gasoline is $3.94 (US EIA website, April 6, 2012). The
US EIA updates its estimates of average gas prices on
a weekly basis. Assume the standard deviation is
$. 25 for the price of a gallon of regular gasoline and
recommend the appropriate sample size for the US
EIA to use if they wish to report each of the following
margins of error at 95% confidence.
a. The desired margin of error is $. 10.
b. The desired margin of error is $. 07.
c. The desired margin of error is $. 05.