Available with MyStatLab™ for Your

Business Statistics Courses

MyStatLab is the market-leading online learning management program for learning
and teaching business statistics.

Statistical Software Support

Built-in tutorial videos and functionality
make using the most popular software
solutions seamless and intuitive. Tutorial
videos, study cards, and manuals
(for select titles) are available within
MyStatLab and accessible at the point
of use. Easily launch exercise and eText
data sets into Excel or StatCrunch, or
copy and paste into any other software
program.

Leverage the Power of StatCrunch

MyStatLab leverages the power of StatCrunch
–powerful, web-based statistical software.
In addition, access to the full online community allows
users to take advantage of a wide variety of resources
and applications at www.statcrunch.com.

Bring Statistics to Life

Virtually flip coins, roll dice, draw cards,
and interact with animations on your
mobile device with the extensive menu
of experiments and applets in
StatCrunch. Offering a number of ways
to practice resampling procedures,
such as permutation tests and bootstrap
confidence intervals, StatCrunch is a
complete and modern solution.

www.mystatlab.com

CAses
Managing Ashland Multicomm
Services
The AMS technical services department has embarked on a
quality improvement effort. Its first project relates to maintain-
ing the target upload speed for its Internet service subscribers.
Upload speeds are measured on a standard scale in which the
target value is 1.0. Data collected over the past year indicate
that the upload speed is approximately normally distributed,
with a mean of 1.005 and a standard deviation of 0.10. Each
day, one upload speed is measured. The upload speed is con-
sidered acceptable if the measurement on the standard scale is
between 0.95 and 1.05.
1. Assuming that the distribution of upload speed has not
changed from what it was in the past year, what is the
probability that the upload speed is
a. less than 1.0?
b. between 0.95 and 1.0?
c. between 1.0 and 1.05?
d. less than 0.95 or greater than 1.05?
2. The objective of the operations team is to reduce the prob-
ability that the upload speed is below 1.0. Should the team
focus on process improvement that increases the mean
upload speed to 1.05 or on process improvement that reduces
the standard deviation of the upload speed to 0.075? Explain.
cardiogood Fitness
Return to the CardioGood Fitness case (stored in
CardioGood Fitness ) first presented on page 52.
1. For each CardioGood Fitness treadmill product line, deter-
mine whether the age, income, usage, and the number of
miles the customer expects to walk/run each week can be
approximated by the normal distribution.
2. Write a report to be presented to the management of Cardio-
Good Fitness detailing your findings.
More Descriptive choices Follow-up
Follow up the More Descriptive Choices Revisited Using
Statistics scenario on page 160 by constructing normal prob-
ability plots for the 3-year return percentages, 5-year return
Cases 237
CHAPTER
6
percentages, and 10-year return percentages for the sample of
407 retirement funds stored in Retirement Funds . In your anal-
ysis, examine differences between the growth and value funds
as well as the differences among the small, mid-cap, and large
market cap funds.
clear Mountain State Student Survey
The Student News Service at Clear Mountain State University
(CMSU) has decided to gather data about the undergraduate
students who attend CMSU. They create and distribute a survey
of 14 questions and receive responses from 111 undergraduates
(stored in StudentSurvey ). For each numerical variable in the
survey, decide whether the variable is approximately normally
distributed by
a. comparing data characteristics to theoretical properties.
b. constructing a normal probability plot.
c. writing a report summarizing your conclusions.
Digital case
Apply your knowledge about the normal distribution in this
Digital Case, which extends the Using Statistics scenario from
this chapter.
To satisfy concerns of potential customers, the management
of MyTVLab has undertaken a research project to learn how
much time it takes users to load a complex video features
page. The research team has collected data and has made some
claims based on the assertion that the data follow a normal
distribution.
Open MTL_QRTStudy.pdf, which documents the work
of a quality response team at MyTVLab. Read the internal
report that documents the work of the team and their conclu-
sions. Then answer the following:
1. Can the collected data be approximated by the normal dis-
tribution?
2. Review and evaluate the conclusions made by the MyTV-
Lab research team. Which conclusions are correct? Which
ones are incorrect?
3. If MyTVLab could improve the mean time by 5 seconds,
how would the probabilities change?

exCeL GUIDE
EG6.1 CONTINUOUS PROBABILITY
DISTRIBUTIONS
There are no Excel Guide instructions for this section.
EG6.2 THE NORMAL DISTRIBUTION
Key Technique Use the NORM.DIST(X value, mean,
standard deviation, True) function to compute normal
probabilities and use the NORM.S.INV(percentage) func-
tion and the STANDARDIZE function (see Section EG3.2)
to compute the Z value.
Example Compute the normal probabilities for Examples
6.1 through 6.3 on pages 220 and 221 and the X and Z val-
ues for Examples 6.4 and 6.5 on pages 223 and 224.
PHStat Use Normal.
For the example, select PHStat ➔ Probability & Prob.
Distributions ➔ Normal. In this procedure’s dialog box
(shown below):
1. Enter 7 as the Mean and 2 as the Standard Deviation.
2. Check Probability for: X 6 = and enter 7 in its box.
3. Check Probability for: X 7 and enter 9 in its box.
4. Check Probability for range and enter 5 in the first
box and 9 in the second box.
5. Check X for Cumulative Percentage and enter 10 in
its box.
6. Check X Values for Percentage and enter 95 in its
box.
7. Enter a Title and click OK.
worksheet Use the COMPUTE worksheet of the
Normal workbook as a template.
The worksheet already contains the data for solving the
problems in Examples 6.1 through 6.5. For other problems,
change the values for the Mean, Standard Deviation, X
238
CHAPTER
3
6
Value, From X Value, To X Value, Cumulative Percent-
age, and/or Percentage.
Unlike most other Excel Guide COMPUTE worksheets,
this worksheet uses formulas in column A to dynamically
create labels based on the data values you enter. These for-
mulas make extensive use of the ampersand operator (&) to
construct the actual label. For example, the cell A10 formula
= "P(X * = "&B8&")" results in the display of P(X 6 =
7) because the initial contents of cell B8, 7, is combined
with "P(X 6 = " and ")". Changing the value in cell B8 to 9,
changes the label in cell A10 to P(X 6 = 9).
EG6.3 EVALUATING NORMALITY
Comparing Data Characteristics
to Theoretical properties
Use the Sections EG3.1 through EG3.3 instructions to com-
pare data characteristics to theoretical properties.
Constructing the Normal probability plot
Key Technique Use an Excel Scatter (X, Y) chart with Z
values computed using the NORM.S.INV function.
Example Construct the normal probability plot for the
one-year return percentages for the sample of 407 retire-
ment funds that is shown in Figure 6.19 on page 230.
PHStat Use Normal Probability Plot.
For the example, open to the DATA worksheet of the
Retirement Funds workbook. Select PHStat ➔ Probabil-
ity & Prob. Distributions ➔ Normal Probability Plot. In
the procedure’s dialog box (shown below):
1. Enter I1:I408 as the Variable Cell Range.
2. Check First cell contains label.
3. Enter a Title and click OK.
In addition to the chart sheet containing the normal prob-
ability plot, the procedure creates a plot data worksheet iden-
tical to the PlotData worksheet discussed in the Worksheet
Excel instructions.

worksheet Use the worksheets of the NPP workbook as
templates.
The NormalPlot chart sheet displays a normal probability
plot using the rank, the proportion, the Z value, and the vari-
able found in the PLOT_DATA worksheet. The PLOT_
DATA worksheet already contains the one-year return
percentages for the example.
To construct a plot for a different variable, paste
the sorted values for that variable in column D of the
PLOT_DATA worksheet. Adjust the number of ranks in
column A and the divisor in the formulas in column B
to compute cumulative percentages to reflect the quantity
n + 1 (408 for the example). (Column C formulas use the
NORM.S.INV function to compute the Z values for those
cumulative percentages.)
exCeL gUiDe 239
If you have fewer than 407 values, delete rows from
the bottom up. If you have more than 407 values, select row
408, right-click, click Insert in the shortcut menu, and copy
down the formulas in columns B and C to the new rows. To
create your own normal probability plot for the 1YrReturn%
variable, open to the PLOT_DATA worksheet and select
the cell range C1:D408. Then select Insert ➔ Scatter (X,
Y) or Bubble Chart (#5 in the labeled Charts groups) and
select the Scatter gallery item. (OS X Excel labels the #5
icon as X Y (Scatter)).
Relocate the chart to a chart sheet, turn off the chart leg-
end and gridlines, add axis titles, and modify the chart title.
 7
CoNTeNTs
Sampling Distributions

“Sampling Oxford
Cereals”

7.1 Sampling Distributions

7.2 Sampling Distribution
of the Mean

VISUAL ExPLORATIONS:
Exploring Sampling
Distributions

7.3 Sampling Distribution

of the Proportion

USING STATISTICS:
Sampling Oxford Cereals,
Revisited

ExCEL GUIDE

objeCTiVes
■ Learn about the con-
cept of the sampling
distribution
■ Compute probabilities
related to the sample
mean and the sample
proportion
■ Understand the impor-
tance of the Central Limit
Theorem
▼
USING sTATisTiCs
Sampling Oxford Cereals

T
he automated production line at the Oxford Cereals main plant fills thousands of boxes
of cereal during each shift. As the plant operations manager, you are responsible for
monitoring the amount of cereal placed in each box. To be consistent with package
labeling, boxes should contain a mean of 368 grams of cereal. Because of the speed of the pro-
cess, the cereal weight varies from box to box, causing some boxes to be underfilled and others
to be overfilled. If the automated process fails to work as intended, the mean weight in the
boxes could vary too much from the label weight of 368 grams to be acceptable.
Because weighing every single box is too time-consuming, costly, and inefficient, you
must take a sample of boxes. For each sample you select, you plan to weigh the individual
boxes and calculate a sample mean. You need to determine the probability that such a sam-
ple mean could have been randomly selected from a population whose mean is 368 grams.
Based on your analysis, you will have to decide whether to maintain, alter, or shut down the
cereal-filling process.
240
 7.2 Sampling Distribution of the Mean 241

i
n Chapter 6, you used the normal distribution to study the distribution of load times for a
MyTVLab web page. In this chapter, you need to make a decision about a cereal-filling
process, based on the weights of a sample of cereal boxes packaged at Oxford Cereals.
You will learn about sampling distributions and how to use them to solve business problems.

7.1 Sampling Distributions

In many applications, you want to make inferences that are based on statistics calculated
from samples to estimate the values of population parameters. In the next two sections, you
will learn about how the sample mean (a statistic) is used to estimate the population mean (a
parameter) and how the sample proportion (a statistic) is used to estimate the population pro-
portion (a parameter). Your main concern when making a statistical inference is reaching con-
clusions about a population, not about a sample. For example, a political pollster is interested
in the sample results only as a way of estimating the actual proportion of the votes that each
candidate will receive from the population of voters. Likewise, as plant operations manager for
Oxford Cereals, you are only interested in using the mean weight calculated from a sample of
cereal boxes to estimate the mean weight of a population of boxes.
In practice, you select a single random sample of a predetermined size from the popula-
tion. Hypothetically, to use the sample statistic to estimate the population parameter, you could
examine every possible sample of a given size that could occur. A sampling distribution is the
distribution of the results if you actually selected all possible samples. The single result you
obtain in practice is just one of the results in the sampling distribution.

7.2 Sampling Distribution of the Mean

In Chapter 3, several measures of central tendency, including the mean, median, and mode,
were discussed. For several reasons, the mean is the most widely used measure of central
tendency, and the sample mean is often used to estimate the population mean. The sampling
distribution of the mean is the distribution of all possible sample means if you select all
possible samples of a given size.

learn More The Unbiased property of the sample Mean

Learn more about the The sample mean is unbiased because the mean of all the possible sample means (of a given
unbiased property of sample size, n) is equal to the population mean, m. A simple example concerning a population
the sample in the SHORT of four administrative assistants demonstrates this property. Each assistant is asked to apply the
TAKES for chapter 7. same set of updates to a human resources database. Table 7.1 presents the number of errors made
by each of the administrative assistants. This population distribution is shown in Figure 7.1.

TABLE 7.1
Number of Errors Administrative Assistant Number of Errors
Made by Each of Ann X1 = 3
Four Administrative Bob X2 = 2
Assistants
Carla X3 = 1
Dave X4 = 4

FIGURE 7.1 3
Number of errors made
Frequency

by a population of four 2
administrative assistants
1

0
0 1 2 3 4
Number of Errors