CHAPTER 8 ---- INTERVAL ESTIMATION
Margin of Error and the Interval Estimate
A point estimator cannot be expected to provide the exact value of the population parameter.
An interval estimate can be computed by adding and subtracting a margin of error to the point estimate.
Point Estimate +/- Margin of Error
The purpose of an interval estimate is to provide information about how close the point estimate is to
the value of the parameter.
The general form of an interval estimate of a population mean is --- x + Margin of Error
Interval Estimate of a Population Mean: s Known
In order to develop an interval estimate of a population mean, the margin of error must be computed
using either:
the population standard deviation s, or
the sample standard deviation s
s is rarely known exactly, but often a good estimate can be obtained based on historical data or other
information.
We refer to such cases as the s known case.
σ
Interval Estimate of m ------- x ± z α / 2 √ n
where: x is the sample mean
1 - a is the confidence coefficient
za/2 is the z value providing an area of a/2 in the upper tail of the standard normal probability
distribution
s is the population standard deviation
n is the sample size
Values of za/2 for the Most Commonly Used Confidence Levels
Meaning of Confidence
Because 90% of all the intervals
constructed using x + 1.645 σ xwill
contain the population mean, we say
we are 90% confident that the interval x
+1.645 σ x includes the population mean m.
We say that this interval has been established at the 90% confidence level.
The value .90 is referred to as the confidence coefficient.
Example: Lloyds Department store
Each week Lloyds department store selects a simple random sample of 100 customers in order to learn
about the amount spent per shopping trip. The historical data indicates that the population follows a normal
distribution.
During most recent week, Lloyd’s surveyed 100 customers (n = 100) and obtained a sample mean of x
= $82. Based on historical data, Lloyd’s now assumes a known value of 𝜎 = $20. The confidence coefficient to
be used in the interval estimate is .95.
95% of the sample means that can be observed are within + 1.96 σ x of the population mean m. The
margin of error is: z α / 2
σ
√n
=1.96
( √20100 )=3.92
Interval estimate of m is:
$82 + $ 3.92 or $78.08 to $85.29
We are 95% confident that the interval contains the population mean.
Using Excel to construct a confidence interval - m Known
Example: Lloyds Department store
In order to have a higher degree of confidence, the margin
of error and thus the width of the confidence interval must
be larger.
Adequate Sample Size
In most applications, a sample size of n ≥ 30 is adequate.
If the population distribution is highly skewed or contains outliers, a sample size of 50 or more is
recommended.
If the population is not normally distributed but is roughly symmetric, a sample size as small as 15 will
suffice.
If the population is believed to be at least approximately normal, a sample size of less than 15 can be
used.
If an estimate of the population standard deviation s cannot be developed prior to sampling, we use the
sample standard deviation s to estimate s.
This is the s unknown case.
In this case, the interval estimate for m is based on the t distribution.
(We’ll assume for now that the population is normally distributed.)
t Distribution
William Gosset, writing under the name “Student”, is the founder of the t distribution.
Gosset was an Oxford graduate in mathematics and worked for the Guinness Brewery in Dublin.
He developed the t distribution while working on small-scale materials and temperature experiments.
The t distribution is a family of similar probability distributions.
A specific t distribution depends on a parameter known as the degrees of freedom.
Degrees of freedom refer to the number of independent pieces of information that go into the
computation of s.
A t distribution with more degrees of freedom has less dispersion.
As the degrees of freedom increase, the difference between the t distribution and the standard normal
probability distribution becomes smaller and smaller.
Comparison of the standard normal distribution with t distributions having 10 and 20 degrees of freedom.
For more than 100 degrees of freedom,
the standard normal z value provides a
good approximation to the t value.
The standard normal z values can be
found in the infinite degrees row (
labeled ∞ ) of the t distribution table.
.
Interval Estimate of a Population Mean: s Unknown
s
x ± tα/2
√n
where: x = the sample mean
1 - a = the confidence coefficient
ta/2 = the t value providing an area of a/2 in the upper tail of a t distribution with n - 1 degrees of
freedom
s = the sample standard deviation
n = the sample size
Example: Credit card debt for the population of US households
The credit card balances of a sample of 70 households provided a mean credit card debt of $9312 with
a sample standard deviation of $4007.
Let us provide a 95% confidence interval estimate of the mean credit card debt for the population of US
households. We will assume this population to be normally distributed.
Interval Estimate of a Population Mean: s Unknown
At 95% confidence, = .05, and /2 = .025.
t.025 is based on n - 1 = 70 - 1 = 69 degrees of freedom.
Interval Estimate of a Population Mean: s Unknown
Example: Credit card debt for the population of US households
�𝑥�� ± 𝑡�.025��𝑠���𝑛��
9312 + 1.995 �4007���70�� = 9312 + 955
We are 95% confident that the mean credit card debt for the population of US households is
between $8357 and $10267.
Using Excel’s Descriptive Statistics Tool
Steps
Step 1: Click the Data tab on the Ribbon
Step 2: In the Analysis group click Data Analysis
Step 3: Choose Descriptive Statistics from the list of Analysis tools
Using Excel’s Descriptive Statistics Tool
Step 4: When the Descriptive statistics dialog box appears
Enter Input Range
Select Grouped by columns
Select Labels in the first row
Select Output range:
Enter C1 in the output range box
Select summary statistics
Select confidence level for mean
Enter 95 in the confidence level for mean box
Click OK
Using Excel’s Descriptive Statistics Tool
Excel Worksheets
95% confidence interval
for credit card balances.
Interval Estimate of a Population Mean: s Unknown
Adequate Sample Size
Usually, a sample size of n ≥ 30 is adequate when using the expression �𝑥�±�𝑡�𝛼/2�𝑠/��𝑛� to develop
an interval estimate of a population mean.
If the population distribution is highly skewed or contains outliers, a sample size of 50 or more is
recommended.
If the population is not normally distributed but is roughly symmetric, a sample size as small as 15 will suffice.
If the population is believed to be at least approximately normal, a sample size of less than 15 can be used.
Summary of Interval Estimation Procedures
for a Population Mean
Sample Size for an Interval Estimate of a Population Mean
Let E = the desired margin of error.
E is the amount added to and subtracted from the point estimate to obtain an interval estimate.
If a desired margin of error is selected prior to sampling, the sample size necessary to satisfy the margin of
error can be determined.
Sample Size for an Interval Estimate of a Population Mean
Margin of Error
𝐸=�𝑧�𝛼/2��𝜎���𝑛��
Necessary Sample Size
n = �(�𝑧�𝛼/2��)�2��𝜎�2���𝐸�2��
Sample Size for an Interval Estimate of a Population Mean
The Necessary Sample Size equation requires a value for the population standard deviation s .
If s is unknown, a preliminary or planning value for s can be used in the equation.
• Use the estimate of the population standard deviation computed in a previous
study.
• Use a pilot study to select a preliminary study and use the sample standard deviation from the
study.
3. Use judgment or a “best guess” for the value of s .
Sample Size for an Interval Estimate of a Population Mean
Example: Cost of renting Automobiles in United States
A previous study that investigated the cost of renting automobiles in the United States found a mean cost of
approximately $55 per day for renting a midsize automobile with a standard deviation of $9.65.
Suppose the project director wants an estimate of the population mean daily rental cost such that there is a .95
probability that the sampling error is $2 or less.
How large a sample size is needed to meet the required precision?
Sample Size for an Interval Estimate of a Population Mean
Example: Cost of renting Automobiles in United States
�𝐸=𝑧�𝛼/2��𝜎���𝑛��=2
At 95% confidence, z.025 = 1.96. Recall that = 9.65.
𝑛=�(1.96�)�2�(9.65�)�2��(2�)�2��=89.43⋍ 90
The sample size needs to be at least 90 mid size automobile rentals in order to satisfy the project director’s $2
margin-of-error requirement.
Interval Estimate of a Population Proportion
The general form of an interval estimate of a population proportion is:
�𝑝� + Margin of Error
Interval Estimate of a Population Proportion
The sampling distribution of �𝑝� plays a key role in computing the margin of error for this interval estimate.
The sampling distribution of �𝑝� can be approximated by a normal distribution whenever np > 5 and n(1 – p)
> 5.
Interval Estimate of a Population Proportion
Normal Approximation of Sampling Distribution of �𝑝�
Interval Estimate of a Population Proportion
�𝑝�±�𝑧�𝛼/2���� �� �𝑝�(1−�𝑝�)�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 normal
probability distribution, and
�𝑝� is the sample proportion
Interval Estimate of a Population Proportion
Example: Survey of women golfers
A national survey of 900 women golfers was conducted to learn how women golfers view their treatment at golf
courses in United States. The survey found that 396 of the women golfers were satisfied with the availability of
tee times.
Suppose one wants to develop a 95% confidence interval estimate for the proportion of the population of
women golfers satisfied with the availability of tee times.
Interval Estimate of a Population Proportion
Example: Survey of women golfers
�𝑝�±�𝑧�𝛼/2���� �� �𝑝�(1−�𝑝�)�𝑛��
where: n = 900, �𝑝� = 396/900 = .44, z/2 = 1.96
.44±1.96��
��.44(1−.44)�900�� = .44 ± .0324
Survey results enable us to state with 95% confidence that between 40.76% and 47.24% of all women golfers
are satisfied with the availability of tee times.
Using Excel to construct a confidence interval
Excel Formula and Value Worksheet
Sample Size for an Interval Estimate of
a Population Proportion
Margin of Error
E = �𝑧�𝛼/2���� �� �𝑝�(1−�𝑝�)�𝑛��
Solving for the necessary sample size n, we get
𝑛=����𝑧�𝛼/2���2��𝑝��1−�𝑝����𝐸�2��
However, �𝑝� will not be known until after we have selected the sample. We will use the planning value p*
for �𝑝�.
Sample Size for an Interval Estimate of
a Population Proportion
Necessary Sample Size
𝑛=����𝑧�𝛼/2���2��𝑝�∗��1−�𝑝�∗����𝐸�2��
The planning value p* can be chosen by:
1. Using the sample proportion from a previous sample of the same or similar units, or
2. Selecting a preliminary sample and using the sample proportion from this sample.
3. Using judgment or a “best guess” for a p* value.
4. Otherwise, using .50 as the p* value.
Sample Size for an Interval Estimate of
a Population Proportion
Example: Survey of women golfers
Suppose the survey director wants to estimate the population proportion with a margin of error of .025 at 95%
confidence.
How large a sample size is needed to meet the required precision? (A previous sample of similar units
yielded .44 for the sample proportion.)
Sample Size for an Interval Estimate of
a Population Proportion
Example: Survey of women golfers
E = �𝑧�𝛼/2���� 𝑝�∗�(1−�𝑝�∗�)�𝑛��= .025
�� �
At 95% confidence, z.0125 = 1.96. Recall that p* = .44.
𝑛=����𝑧�𝛼/2���2��𝑝�∗��1−�𝑝�∗����𝐸�2��=���1.96��2�(.44)�.56���(.025)�2��=
1514.5
A sample of size 1515 is needed to reach a desired precision of + .025 at 95% confidence.
Sample Size for an Interval Estimate of
a Population Proportion
Note: We used .44 as the best estimate of p in the preceding 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 1537.
Implications of Big Data
As the sample size becomes extremely large, the margin of error becomes extremely small and resulting
confidence intervals become extremely narrow.
No interval estimate will accurately reflect the parameter being estimated unless the sample is relatively free of
nonsampling error.
Statistical inference along with information collected from other sources can help in making the most informed
decision.
End of Chapter 8
