Applied Statistics in Business and

Economics, 7e ISE - eBook PDF

Visit to download the full and correct content document:
https://ebooksecure.com/download/applied-statistics-in-business-and-economics-7e-i
se-ebook-pdf/
Applied Statistics
in Business and Economics
Seventh Edition
 Final PDF to printer

The McGraw Hill/Irwin Series in Operations and Decision Sciences

SUPPLY CHAIN MANAGEMENT BUSINESS SYSTEMS DYNAMICS Slater and Wittry

Math for Business and Finance: An Algebraic
Bowersox, Closs, Cooper, and Bowersox Sterman Approach
Supply Chain Logistics Management Business Dynamics: Systems Thinking and Second Edition
Fifth Edition ­Modeling for a Complex World
First Edition BUSINESS STATISTICS
Johnson
Purchasing and Supply Management OPERATIONS MANAGEMENT Bowerman, Drougas, Duckworth, Froelich,
Sixteenth Edition Hummel, Moninger, and Schur
Cachon and Terwiesch Business Statistics in Practice
Simchi-Levi, Kaminsky, and Simchi-Levi Operations Management Ninth Edition
Designing and Managing the Supply Chain: Second Edition
­Concepts, Strategies, Case Studies Doane and Seward
Fourth Edition Cachon and Terwiesch Applied Statistics in Business and Economics
Matching Supply with Demand: An Introduction Seventh Edition
PROJECT MANAGEMENT to Operations Management
Fourth Edition Doane and Seward
Larson and Gray Essential Statistics in Business and Economics
Project Management: The Managerial Process Jacobs and Chase Third Edition
Eighth Edition Operations and Supply Chain Management
Sixteenth Edition Lind, Marchal, and Wathen
SERVICE OPERATIONS MANAGEMENT Basic Statistics for Business and Economics
Jacobs and Chase Tenth Edition
Bordoloi, Fitzsimmons, and Fitzsimmons Operations and Supply Chain Management:
Service Management: Operations, Strategy, Infor- The Core Lind, Marchal, and Wathen
mation Technology Fifth Edition Statistical Techniques in Business and Economics
Ninth Edition Eighteenth Edition
Schroeder and Goldstein
MANAGEMENT SCIENCE Operations Management in the Supply Chain: Jaggia and Kelly
Decisions and Cases Business Statistics: Communicating with Numbers
Hillier and Hillier Eighth Edition Fourth Edition
Introduction to Management Science: A Modeling
and Case Studies Approach with Spreadsheets Stevenson Jaggia and Kelly
Sixth Edition Operations Management Essentials of Business Statistics: Communicating
Thirteenth Edition with Numbers
BUSINESS RESEARCH METHODS
Second Edition
Swink, Melnyk, and Hartley
Schindler
Managing Operations Across the Supply Chain BUSINESS ANALYTICS
Business Research Methods
Fourth Edition
Fourteenth Edition
Jaggia, Kelly, Lertwachara, and Chen
BUSINESS MATH Business Analytics: Communicating with Numbers
BUSINESS FORECASTING
First Edition
Slater and Wittry
Keating and Wilson
Practical Business Math Procedures
Business Forecasting
Thirteenth Edition
Seventh Edition

doa16287_fm_i-1.indd ii 10/21/20 10:22 AM

Applied Statistics
in Business and Economics
Seventh Edition

David P. Doane
Oakland University

Lori E. Seward
University of Colorado
 Final PDF to printer

APPLIED STATISTICS
Published by McGraw Hill LLC, 1325 Avenue of the Americas, New York, NY 10121. Copyright © 2022 by
McGraw Hill LLC. All rights reserved. Printed in the United States of America. No part of this publication may
be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without
the prior written consent of McGraw Hill LLC, including, but not limited to, in any network or other electronic
storage or transmission, or broadcast for distance learning.
Some ancillaries, including electronic and print components, may not be available to customers outside the
United States.
This book is printed on acid-free paper.
1 2 3 4 5 6 7 8 9 LWI 24 23 22 21

ISBN 978-1-260-59764-6
MHID 1-260-59764-4

Cover Image: Xuanya Han/Moment/Getty Images

All credits appearing on page or at the end of the book are considered to be an extension of the copyright page.

The Internet addresses listed in the text were accurate at the time of publication. The inclusion of a website does
not indicate an endorsement by the authors or McGraw Hill LLC, and McGraw Hill LLC does not guarantee the
accuracy of the information presented at these sites.

mheducation.com/highered

doa97644_fm_ise.indd iv 10/10/20 09:14 AM

About the Authors
Courtesy of David Doane
Courtesy of Lori Seward
Dedication
To all my students who challenged me to make statistics
David P. Doane
David P. Doane is accredited by the American Statistical Association as a Professional Statis-
tician (PStat®) . He is professor emeritus in Oakland University’s Department of Decision and
Information Sciences. He earned his Bachelor of Arts degree in mathematics and economics
at the University of Kansas and his PhD from Purdue University’s Krannert Graduate School.
His research and teaching interests include applied statistics, forecasting, and statistical edu-
cation. He is co-recipient of three National Science Foundation grants to develop software to
teach statistics and to create a computer classroom. He is a longtime member of the American
Statistical Association, serving in 2002 as president of the Detroit ASA. He has consulted
with governments, health care organizations, and local firms. He has published articles in
many academic journals. He currently belongs to ASA chapters in San Diego and Orange
County/Long Beach.
Lori E. Seward
Lori E. Seward is a teaching professor in The Leeds School of Business at the University
of Colorado in Boulder. She earned her Bachelor of Science and Master of Science degrees
in Industrial Engineering at Virginia Tech. After several years working as a reliability and
quality engineer in the paper and automotive industries, she earned her PhD from Virginia
Tech and joined the faculty at The Leeds School in 1998. Professor Seward has served as the
faculty director of Leeds’ MBA programs since 2017. She currently teaches as well as coor-
dinates the core statistics course for the Leeds full-time, Professional, and Executive MBA
programs. She served as the chair of the INFORMS Teachers’ Workshop for the annual 2004
meeting. Her teaching interests focus on developing pedagogy that uses technology to create
a collaborative learning environment in large undergraduate and MBA statistics courses. Her
most recent article, co-authored with David Doane, was published in the Journal of Statistics
Education (2011).
To Robert Hamilton Doane-Solomon
David
relevant to their lives.
Lori
v
 From the Authors
“How often have you heard people/students say about a particular subject, ‘I’ll never
use this in the real world’? I thought statistics was a bit on the ‘math-geeky’ side at
first. Imagine my horror when I saw α, R2, and correlations on several financial reports
at my current job (an intern position at a financial services company). I realized then
that I had better try to understand some of this stuff.”
—Jill Odette (an introductory statistics student)

As recently as a decade ago our students used to ask us, “How do I use statistics?” Today we
more often hear, “Why should I use statistics?” Applied Statistics in Business and Economics
has attempted to provide real meaning to the use of statistics in our world by using real busi-
ness situations and real data and appealing to your need to know why rather than just how.
With over 50 years of teaching statistics between the two of us, we feel we have something
to offer. Seeing how students have changed as the new century unfolds has required us to
adapt and seek out better ways of instruction. So we wrote Applied Statistics in Business and
Economics to meet four distinct objectives.
Objective 1: Communicate the Meaning of Variation in a Business Context Variation
exists everywhere in the world around us. Successful businesses know how to measure varia-
tion. They also know how to tell when variation should be responded to and when it should be
left alone. We’ll show how businesses do this.
Objective 2: Use Real Data and Real Business Applications Examples, case studies, and
problems are taken from published research or real applications whenever possible. Hypo-
thetical data are used when it seems the best way to illustrate a concept.
Objective 3: Incorporate Current Statistical Practices and Offer Practical Advice
With the increased reliance on computers, statistics practitioners have changed the way they
use statistical tools. We’ll show the current practices and explain why they are used the way
they are. We also will tell you when each technique should not be used.
Objective 4: Provide More In-Depth Explanation of the Why and Let the Software
Take Care of the How It is critical to understand the importance of communicating with
data. Today’s computer capabilities make it much easier to summarize and display data than
ever before. We demonstrate easily mastered software techniques using the common software
available. We also spend a great deal of time on the idea that there are risks in decision mak-
ing and those risks should be quantified and directly considered in every business decision.
Our experience tells us that students want to be given credit for the experience they bring
to the college classroom. We have tried to honor this by choosing examples and exercises set
in situations that will draw on students’ already vast knowledge of the world and knowledge
gained from other classes. Emphasis is on thinking about data, choosing appropriate analytic
tools, using computers effectively, and recognizing limitations of statistics.

What’s New in This Seventh Edition?

In this edition we have listened to you and have made many changes that you asked for. We
sought advice from students and faculty who are currently using the textbook and reviewers at
a variety of colleges and universities. At the end of this preface is a detailed list of chapter-by-
chapter improvements, but here are just a few of them:
∙ New, brighter overall design (format, colors).
∙ Continued strong focus on Excel and business applications.
∙ New Analytics in Action briefings on careers in data analytics and exciting applications of
big data, artificial intelligence, and machine learning (including ethical issues).
∙ New Appendix J with side-by-side comparison of statistics functions in Excel and R.
∙ New Appendix K with an easy walk-through to get started with R and RStudio.

vi
∙ Chapter-end Software Supplements showing how to use R for applications in that chapter.
∙ Updated exercises with emphasis on compatibility with Connect®.
∙ Updated test bank questions matched with topics and learning objectives.
∙ New and updated Mini Cases for economics and business.
∙ New and updated exercise data sets, web links, Big Data Sets, and Related Reading.
∙ Many new guided examples on Connect. Students can watch 90 guided examples to aid
their learning.
∙ Connect® supplements including LearningStats demonstrations, illustrations of R calcula-
tions for common tasks, and video tutorials (both PC and Mac).

Software
Excel is used throughout this book because it is available everywhere. Some calculations are
illustrated using MegaStat and Minitab because they offer more capability than Excel’s Data
Analysis Tools. In recognition of growing interest in analytics training beyond Excel, our
textbook now provides an optional introduction to R with illustrations of topics in each chap-
ter. Our support for R is further enhanced with LearningStats modules, tables of R functions,
and R-compatible Excel data sets. To further assist students we provide Connect® tutorials or
demonstrations on using Excel, Minitab, MegaStat, and R. At the end of each chapter is a list
of LearningStats demonstrations that illustrate the concepts from the chapter.

Math Level
The assumed level of mathematics is pre-calculus, though there are rare references to calculus
where it might help the better-trained reader. All but the simplest proofs and derivations are
omitted, though key assumptions are stated clearly. The learner is advised what to do when
these assumptions are not fulfilled. Worked examples are included for basic calculations, but
the textbook does assume that computers will do the calculations after the statistics class is
over, so interpretation is paramount. End-of-chapter references and suggested websites are
given so that interested readers can deepen their understanding.

Exercises
Simple practice exercises are placed within each section. End-of-chapter exercises tend to be
more integrative or to be embedded in more realistic contexts. Attention has been given to
revising exercises so that they have clear-cut answers that are matched to specific learning
objectives. A few exercises invite short answers rather than just quoting a formula. Answers to
odd-numbered exercises are in the back of the book (all of the answers are in the instructor’s
manual).

LearningStats
Connect users can access LearningStats, a collection of Excel spreadsheets, Word documents,
and PowerPoints for each chapter. It is intended to let students explore data and concepts at
their own pace, ignoring material they already know and focusing on things that interest them.
LearningStats includes deeper explanations on topics such as how to write effective reports,
how to perform calculations, or how to make effective charts. It also includes topics that did
not appear prominently in the textbook (e.g., partial F test, Durbin–Watson test, sign test,
bootstrap simulation, and logistic regression). Instructors can use LearningStats PowerPoint
presentations in the classroom, but Connect users also can use them for self-instruction. No
instructor can “cover everything,” but students can be encouraged to explore LearningStats
data sets and/or demonstrations, perhaps with an instructor’s guidance.
David P. Doane
Lori E. Seward

vii
 1
CHAPTER Chapter 4 Descriptive Statist

How Are Chapters Organized

Overview of Square Selling Price Square Selling Price Square Selling Price

Statistics
Feet (thousands) Feet (thousands) Feet (thousands)
3,570 861 3,240 809 3,160 778
3,410 740 2,660 639 3,310 760
2,690 563 3,160 778 2,930 729
Chapter Contents 3,260
3,130
698
624
3,460
3,340
737
806
3,020
2,320
720
575
Each chapter begins with a short list of CHAPTER
3,460 CONTENTS
737 3,240 809 3,130 785
section topics that are covered in the 3,340 8061.1 What Is Statistics? 2,660 639
1.2 Why Study Statistics?
chapter. 1.3 Applying Statistics in Business
1.4 Statistical Challenges
1.5 Critical Thinking

Chapter Learning Objectives GROUPED DATA

CHAPTER LEARNING OBJECTIVES
4.7
Each chapter includes a list of learn- When you finish this chapter, you should be able to

ing objectives students should be able Weighted Mean

LO 1-1
LO 1-2
Define statistics and explain some of its uses.
List reasons for a business student to study statistics.
LO 4-10
to attain upon reading and studying the The weighted mean is a Explain
LO 1-3 sum that assigns
the uses of statisticseach data value a weight wj that represents a frac-
in business. Calculate the mean
chapter material. Learning objectives tion of the total (i.e., the kState
LO 1-4 weights must
the common sum facing
challenges to 1).business professionals using statistics. standard deviation f
First Pages
grouped data.
give students an overview of what is LO 1-5 List and explain common statistical pitfalls.
_ k k
expected and identify the goals for learn- x = ∑ wj xj Where ∑ wj = 1.00 (4.25)
j=1 j=1
ing. Learning objectives also appear
next to chapter topics in the margins. 60
For Applied Statistics in Business and Economics
example, your instructor might give a weight of 30 percent to homework, 20 percent
to the midterm
Section Exercises exam,3.140 (a)percent to the final
Make a stem-and-leaf plot for exam, and 10 on
these 24 observations percent toof acustomers
the number term who project
used a
(so that .30 + .20 + .40 + .10 = 1.00).
downtown Suppose
Citibank ATM during your scores
the noon hour on
on 24these wereworkdays.
consecutive 85, 68, (b)78,
Makeand a dot90.
plot

Section Exercises Your weighted average for the Citibank

®
of the ATM data. (c) Describe these two displays. (Hint: Refer to center, variability, and shape.)
course would be
Multiple section exercises are found _ k
39 32 21 26 19 27 32 25
x = ∑ wj xj = .30
18 × 85
26+ .20
34× 6818+ .4031× 78 35
+ .10 21× 90 =
33 79.3
throughout the chapter so that students j=1 33 9 16 32 35 42 15 24

can focus on material just learned. Despite a low score on3.2the (a)
midterm exam, you
Make a stem-and-leaf aretheright
plot for numberat
of the borderline
defects forforan
per 100 vehicles 8032 (if
these your
brands. (b)
Make a dot plot of the defects data. (c) Describe these two displays. (Hint: Refer to center, vari-
instructor rounds up). The weighted mean is widely used in cost accounting (weights for cost
ability, and shape.)
categories), finance (asset weights in investment portfolios), and other business applications.
2
Defects per 100 Vehicles (alphabetical by brand) JDPower

Mini Case 4.8

Brand Defects Brand Defects Brand Defects
Acura 86 Hyundai 102 Confirming Pages
MINI 133
doa16287_ch01_002-023.indd 2 Audi 111 Infiniti 107 05/14/20 Mitsubishi
07:59 AM 146
What Is the DJIA? DJIA
BMW 113 Jaguar 130 Nissan 111
Buick 114 Jeep 129 Porsche 83
The Dow Jones Industrial Cadillac
Average (commonly
111 called
Kia the DJIA) is
126 the oldest
Ram U.S. stock
110
market price index, based on the prices111
Chevrolet of 30 large, widely held, 170
Land Rover and actively Scion traded “blue 114
chip” public companies inChrysler
the United 122 Lexus
States (e.g., 88
Coca-Cola, Microsoft, Subaru
Walmart, Walt121
326 Applied Statistics in Business and Economics Dodge 130 Lincoln 106 Toyota 117
Mini Cases Disney). Actually, only a few
Fordof its 30 component
93 Mazdacompanies are 114“industrial.”VolkswagenThe DJIA135is
not a simple average but rather
GMC a weighted126average based on prices 87
Mercedes-Benz of its component
Volvo stocks.
109
Every chapter includes two or three Can a Null Hypothesis Be Proved?
Originally a simple mean Honda
of stock prices,95 the Mercury 113
DJIA now is calculated as the sum of the
mini cases, which are solved applica- 30 stock
No, pricesprove
we cannot divided bySource:
a null ahypothesis—we
“divisor” that compensates
J.D. Power and Associates
can only failfor
2010 Initial Quality Study™.
to stock
rejectsplits it. A andnullother changesthat
hypothesis
tions. They show and illlustrate the over time.
survives The divisor
repeated is revised
tests3.3without as often as“true”
(a) Makerejection
a stacked dotisplot
necessary
to compare only (see
the in the
Theoflimited
lengths cell phonesense
Wall calls by that
Street Sarah it
Journal for
andhas
the
Bob been
during
latest divisor
thoroughly value). Because
scrutinized lasthigh-priced
andthetested. week.Today’s
(b) Describestocks
“true”
what the comprise
dot plots tellayou.
hypothesis larger proportion
couldPhoneCalls
be disproved of the sum,
tomorrow
analytical application of specific statis-
ifthe
newDJIAdataisare
more strongly
found. If weaffected to by
failSarah’s rejectchanges in
H1,01,, 1,the high-priced
same hypothesis stocks.may Thatbeis,retested.
a 10 percent That is
tical concepts at a deeper level than the pricescientific
how increaseinquiry
in a $10 stock Einstein’s
works. would
calls:
have
Bob’s calls: less
theories,
1, 2, 3, 3, 3, 5, 5, 6, 6, 7, 8, 8, 12, 14, 14, 22, 23, 29, 33, 38, 45, 66
5, 8, 9,effect
14,for than
23, 23,a24,10
example,
17, 21, 26,arepercent
27, over
27, price
28, 29,100 increase
31, 33,years
35, 39, 41old but in aare
examples. $50being
still stock,subjected
even if both to companies
rigorous tests. haveFew the same
scientiststotal marketreally capitalization
think that Einstein’s (the totaltheories
numberare
of sharesYet
“wrong.” timesit’sthe
in price per share;
the nature often referred
of science to keeptotrying as “market to refute cap”). Broad-based
accepted theories, market espe-
price when
cially indexes (e.g.,test
a new NSDQ, AMEX,
is possible or NYSE,
when new S&Pdata 500,become Russ 2K) are widely
available. used by
Similarly, thefundsafety
ofmanagers,
commonly butused Mini“Dow”
the venerable
prescription Case drugs 3.1
is isstillcontinually
the one youbeing see first studied.on CNN or MSNBC.
Sometimes, “safe” drugs
are revealed to have serious side effects
U.S. Business Cycles only after large-scale, long-term use by consumers.
Although many businesses anticipated the recession that followed the housing price bub-
ble of the early 2000s, they needed to anticipate its probable length to form strategies for
Analytics in Action debt management and future product releases. Fortunately, good historical data are avail-
able from the National Bureau of Economic Research, which keeps track of business

Analytics in Action
cycles. The length of a contraction is measured from the peak of the previous expansion
Walmart, Big Data, and Retail Analytics
to the beginning of the next expansion based on the real gross domestic product (GDP).
Table 3.3 shows the durations, in months, of 33 U.S. recessions.
These NEW features bring in real- Walmart processes overFrom the dot customer
a million plot in Figure 3.1, we see that
transactions the hour,
each 65-month contraction
which (1873–1879)
translates into
was quite unusual, although four recessions did exceed 30 months. Most recessions have
world examples to illustrate data ana- doa16287_ch04_100-157.indd 141
two to three petabyteslasted
of data each hour. (A petabyte is a million gigabytes!) What to do
less than 20 months. Only 7 of 33 lasted less than 10 months. The 8-month 2001
07/29/20

lytics in action. with all this data? Walmart

recessioniswaspracticing
among the data democratization.
shortest. Although the recessionThisthatterm
beganmeans making
in December 2007
large amounts of datahadavailable to everyone
unique features in the
(major financial organization
crisis, so that employees
steep rise in unemployment, can
sluggish recov-
ery), its duration (18 months) was not extreme.
quickly react to changes in their customers’ behaviors.
Walmart operates a Data Café that can be accessed by everyone in the company.
When store managers notice changes in sales for particular products, they can go to
the Data Café and look at data across all their stores to figure out why the changes
are happening. For example, a sharp decrease in sales of one product turned out to be
viii caused by incorrect pricing for that product. The problem can be fixed within a day
rather than having to wait a week or two—which is what typically happened in the past.
Differences60in sales between stores can highlight mistakes such as Halloween products
doa16287_ch03_056-099.indd 07/28/20 10:24 PM

not making it to the shelf. Identifying the mistake and fixing it in time for customers
to purchase before the big night prevents a loss in profit. Big data mean that real-time
 January 1893 (I) June 1894 (II) 17 April 1960 (II) February 1961 (I) 10
December 1895 (IV) June 1897 (II) 18 December 1969 (IV) November 1970 (IV) 11
June 1899 (III) December 1900 (IV) 18 November 1973 (IV) March 1975 (I) 16
September 1902 (IV) August 1904 (III) 23 January 1980 (I) July 1980 (III) 6

to Promote Student Learning?

May 1907 (II)
January 1910 (I)
January 1913 (I)
June 1908 (II)
January 1912 (IV)
December 1914 (IV)
13
24
23
July 1981 (III)
July 1990 (III)
March 2001 (I)
November 1982 (IV)
March 1991 (I)
November 2001 (IV)
16
8
8
August 1918 (III) March 1919 (I) 7 December 2007 (IV) March 2009 (I) 18
January 1920 (I) July 1921 (III) 18

Source: U.S. Business Contractions found at www.nber.org.

Figures and Tables

The table supplies information that the dot plot cannot. For example, during the 1930s
Throughout the text,
there were actually two there are hundreds
major contractions of charts,
(43 months from graphs, tables,
1929 to 1933, and spreadsheets to illustrate statistical concepts being
13 months
applied.
from 1937 to 1938), which is one reason why that period seemed so terrible to thosetext
These visuals help stimulate student interest and clarify the who explanations.
lived through it. The Great Depression of the 1930s was so named because it lasted a
long time and the economic decline was deeper than in most recessions. Confirming Pages

Figure 3.1
Dot Plot of Business Cycle Duration (n = 33)

Peak-to-Trough Business Cycle Duration

Chapter 2 Data Collection 37

0 10 20 Random
30 Sampling
40 Methods
50 60 70
We will Number
first discuss the four random sampling techniques shown in Table 2.5 and then
of Months
describe three commonly used non-random sampling techniques, summarized in Table 2.8. First Pages

Simple Random Sample Use random numbers to select items from a list (e.g., Visa Table 2.5
cardholders). Random Sampling
3.2 FREQUENCY DISTRIBUTIONS
Systematic Sample
80
Select every kth item from a list or sequence (e.g., restaurant
Applied Statistics in Business and Economics
customers).
Methods

AND HISTOGRAMS Stratified Sample Select randomly within defined strata (e.g., by age,
occupation, 3.7 gender). SCATTER PLOTS
Cluster Sample Select random geographical regions (e.g., zip codes) that
Frequency Distributions LO 3-8 represent Athe scatter plot shows n pairs
population. LOof observations
3-2 (x1, y1), (x2, y2), . . ., (xn, yn) as dots (or some
A frequency distribution is a table formed Make by classifying other symbol) on an X-Y graph. This type of display is so important in statistics that it deserves
and interpretnanumerical data values into k Create a frequency distri-
classes called bins. The table shows the frequency of data values careful attention.
eachA scatter plot is a starting point for bivariate data analysis. We create scatter
by N that fall within bybin.
scatter plot.
We denote the population size and the
plotssample size
to investigate In abution
simple
n.relationship
the
forrandom
a data set.
between twosample,
variables. Typically, we would like to know
Frequencies also can be expressed as relative frequencies or percentages
every item in the population of N items hasifthe of the
same
there
total number
is anchance of being
association chosen
between in the sample
two variables and, if of
so, what kind of association exists. As
of observations. n items. A physical experiment to accomplish we this
did with
wouldunivariate data analysis,
be to write each oflet’s
the look at avalues
N data scatter plot to see what we can observe.
on a poker chip and then to draw n chips from a bowl after stirring it thoroughly. But we can
Examples accomplish the same thing if the N population items appear on a numbered list, by choosing n
Example 3.3
integers between 1 and N that we match up against Figurethe
3.16numbered
shows a scatter plot
list of with
the life expectancy
population items. on the X-axis and birth rates on the
Examples of interest to For stu- Birth Rates
example, suppose and Life
we want studentInatthis
to select oneY-axis. illustration,
random from there
a list seems
of 15 to be an association
students (see between X and Y. That is,
nations with higher birth rates tend to have lower life expectancy (and vice versa). No
dents are taken from publishedFigure 2.5). If youExpectancy
were asked to “use your judgment,” you would probably pick a name in the mid-
cause-and-effect relationship is implied because, in this example, both variables could
research or real applications dle, thereby
to biasing theThedraw
Source: World against those individuals
Factbook 2003. at either
be influenced by aend ofvariable
third the list.that
Instead
is notwe rely on (e.g., GDP per capita).
mentioned
Central Intelligence Agency, 2003.
a random number to “pick” the name. How do we determine the random number? Before com-
illustrate the statistics concept. www.cia.gov.
puters, statisticians relied on published tables ofFigure random 3.16numbers. The process is simpler
doa16287_ch03_056-099.indd 61 07/28/20 today.
10:24 PM
For the most part, examples Most pocket calculators have a key to produce a random decimal in the interval [0, 1] that can be
Scatter Plot of Birth Rates and Life Expectancy (n = 153 nations) BirthLife
are focused on business, but to a random integer. In this example, we used Excel’s function =RANDBETWEEN(1,15) to pick
converted
a random 60
there are also some that are integer between 1 and 15. The number was 12, so Stephanie was selected. There is no
more general and don’t bias because all values from 1 to 15 are equiprobable (i.e., equally50likely to occur). An equivalent
Birth Rate per 1,000

require
R function for choosing a single random integer between 1 and 15 40 is sample(1:15,1,1).
any prerequisite knowledge. 30
And there are some that are Random person 12 20 Figure 2.5
based on student projects. 10
0
Picking on Stephanie
1 Adam 6 Haitham 11 Moira
30 40 50 60 70 80
2 Addie 7 Jackie 12 Stephanie
Life Expectancy (years)
3 Don 8 Judy 13 Stephen
4 Floyd 9 Lindsay 14 Tara
5 Gadis 10 Majda 15 Xander
Figure 3.17 shows some scatter plot patterns similar to those that that you might observe
when you have a sample of (X, Y) data pairs. A scatter plot can convey patterns in data pairs
Sampling without replacement means that that once
wouldannot be has
item apparent
beenfrom a table.
selected Compare
to be includedtheinscatter plots in Figure 3.18 with the
prototypes
the sample, it cannot be considered for the sample and The
again. use your own
Excel words to=RANDBETWEEN(a,b)
function describe the patterns that you see.
uses sampling with replacement. This means that the same random number could show up
more than once. Figure
Using the3.17
bowl analogy, if we throw each chip back in the bowl and stir the
Data Set Icon
Strong Positive Weak Positive No Pattern
contents before the next draw, an item can be chosen again. Instinctively most people believe
Prototype Scatter Plot
that sampling the
A data set icon is used throughout without
text replacement
to identifyisdata
preferred
sets over sampling
used with replacement
in the figures, examples, because
and
Patterns Y Y Y USTrade
allowing duplicates in our sample seems odd. In reality, sampling without replacement can
exercises that are included in Connect for the text.
be a problem when our sample size n is close to our population size N. At some point in the
sampling process, the remaining items in the population will no longer have the same prob-
X X X
ability of being selected as the items we chose at the beginning of the sampling process. This
could lead to a bias (a tendency to overestimate or underestimate
Strong Negative the parameterWeak
we Negative
are trying Nonlinear Pattern
to measure) in our sample results. Sampling with replacement does not lead to bias. In a list of
items to be sampled (a vector x), the R function sample(x, n, 1) will choose a random sample of n
ix
items with replacement or use sample(x, n, 0) toY sample without replacement). Y Y
When should we worry about sampling without replacement? Only when the population is
finite and the sample size is close to the population size. Consider the Russell 3000® Index,
X X X
How Does This Text Reinforce First Pages

First Pages

Chapter 4 Descriptive Statistics 147

Chapter Summary
Chapter Summary Chapter 4 Descriptive Statistics 147
Chapter summaries provide The mean and median describe a sample’s center and also indicate the mean absolute deviation or MAD is easy to understand but
an overview of the material Chapter
skewness. TheSummary
mode is useful for discrete data with a small range. lacks nice mathematical properties. Quartiles are meaningful
The trimmed mean eliminates extreme center
values. and
Thealso
geometric even for fairly small deviation
data sets, or
while is easy toareunderstand
percentiles used only but
for
covered in the chapter. The mean and median describe a sample’s indicate the mean
large
absolute MAD
mean mitigates
skewness. high is
The mode extremes butdiscrete
useful for cannot data
be used
with when zeros
a small or
range. lacks data
nicesets. Box plots show
mathematical the quartiles
properties. and are
Quartiles data meaningful
range. The
negative values are present. The midrange is easy
The trimmed mean eliminates extreme values. The geometric to calculate correlation
even for fairly small datameasures
coefficient sets, whilethepercentiles
degree of linearity
are used between
only for
but
mean is mitigates
sensitive to extremes.
high extremesVariability
but cannotis be
typically measured
used when zeros by
or two
largevariables.
data sets.The plots showmeasures
Boxcovariance the degree
the quartiles and datato range.
which The
two
the standard
negative valuesdeviation, whileThe
are present. relative dispersion
midrange is given
is easy by the
to calculate variables move together. We can estimate many common
correlation coefficient measures the degree of linearity between descriptive
coefficient of variation for nonnegative data. Standardized
but is sensitive to extremes. Variability is typically measured by data statistics from The
two variables. grouped
covariance Sample coefficients
data. measures the degree toFirst Pages
of which
skewness
two
reveal or unusual data
outliersdeviation,
the standard whilevalues, anddispersion
relative the Empirical Ruleby
is given offers
the and kurtosis
variables moveallow moreWe
together. precise inferences
can estimate about
many the shape
common of the
descriptive
acoefficient
comparison of with a normal
variation distribution. data.
for nonnegative In measuring dispersion,
Standardized data population being sampled instead of relying on histograms.
statistics from grouped data. Sample coefficients of skewness
reveal outliers or unusual data values, and the Empirical Rule offers and kurtosis allow more precise inferences about the shape of the
a comparison with a normal distribution. In measuring dispersion, population being sampled instead of relying on histograms.
Key Terms
148 Applied Statistics in Business and Economics
Key Terms Center
Key Terms
geometric mean
Variability
Chebyshev’s Theorem
Shape
bimodal distribution
Other
box plot
Commonly Used Formulas in Descriptive Statistics covariance
Key terms are highlighted mean
Center
median
coefficient of variation
Variability
Empirical Rule
kurtosis
Shape
kurtosis coefficient
Other
five-number summary
geometric mean Chebyshev’s Theorem bimodal distribution box plot
and defined within the text. midhinge
mean mean absolute
coefficient
Sample mean: of¯ deviation
1 n
x variation
= __ ∑ xi
leptokurtic
kurtosis interquartile range
covariance
midrange outliers mesokurtic method of medians
They are also listed at the median Empirical Rule n i =1 kurtosis coefficient
multimodal
five-number summary
quartiles
mode
midhinge population
mean absolutevariance _________
deviation leptokurtic distribution interquartile range
ends of chapters to aid in trimmed
midrange mean Geometric range
mean:
outliers G = √
n
x x
1 2 . . . xn negatively
mesokurticskewed sample
method correlation
of medianscoefficient
weighted mean sample variance Pearson 2 skewness coefficient
reviewing. mode population variance __ multimodal distribution quartiles
standard
range deviation √
platykurtic
xn
n−1 __
trimmed mean Growth rate: GR = −1 negatively skewed sample correlation coefficient
weighted mean standardized
sample variance data x1 positively
Pearson 2 skewed
skewness coefficient
two-sum
Range: Range = xmax − xmin Schield’s
formula
standard deviation platykurticRule
z-score
standardized data skewed left
positively skewed
xmax + xminskewed
Schield’sright
Midrange = _________
two-sum formula
Midrange: Rule First Pages
z-score 2 skewness
skewed left
___________ skewness
skewed coefficient
right

√
n
∑ (x − ¯ x)2 symmetric data
skewness
i
Sample standard deviation: s= ___________
i =1
skewness coefficient
n−1
symmetric data
Choosing the Appropriate Statistic or Visual Display
148
Coefficient of variation: Population
Applied Statistics in Business and Economics
Sample

Commonly Used
σ s
Choosing
CVthe
= 100
Nominal ×_
Appropriate StatisticCV
or =
Visual __
100 ×Display
Ordinal ¯
Commonly Used Formulas in Descriptive
μ
Statistics
Data x
Median
Formulas Mode
StandardizedBar
variable:
Chart Population
Nominal
Type?
Sample
Ordinal
Mode
Data Bar Chart
Column Chart 1x − n Median
Some chapters provide a SampleMode
mean: ¯
zxi =
__ i∑μ
= n_____ xi Type? xi − ¯
zi = _____
x Column Chart
Mode
Bar Chart iσ
=1 Interval or s
listing of commonly used Column Chart _________ Ratio
Bar Chart
Geometric mean: G=√ n
x1 x2 . _______
. . x+ Q
Q Column Chart
formulas for the topic under Midhinge: Midhinge = 1 n 3 Interval or
__ 2
Ratio
discussion. √
Center n−1 __ xn Variability
Growth rate:
Mean GR = nx
− 1Describe Range
Median ∑1(xi − ¯ )(y − ¯
xWhat? y)
sXY Interquartile Range
Range:
Sample correlation coefficient:
Mode = _________________________
Range
rCenter = xi =1 − x
_________max min
_________ or r = ____
Variability s s Standard Deviation
Describe
Mean
√ √ Range
n n
Midrange ∑ (x i − ¯ x ) 2 ∑ (y i − ¯ y)2 X Y
Coefficient of
Median i =1 xmax +What?xmin
i =1 Interquartile Range
Midrange:
Geometric Mean Midrange = _________ Variation
Mode 2 Shape Standard Deviation
Midhinge Standardized z-Values
Midrange k k
Coefficient of
Weighted mean:
Histogram x = ∑___________
¯ n j xj where ∑ wj = 1.00
w Histogram

√
Geometric Mean j = 1∑ (x − ¯ x ) 2 j =Shape1 Variation
Box Plot i Box Plot
Midhinge
Sample standard deviation: s = ___________
i =1 Standardized z-Values
Mean vs. Median
Histogram k f m n − 1 Histogram
Grouped mean: ¯
x=∑ ____
j Skewness
j Coefficient
Box Plot j = 1 n Kurtosis Coefficient Box Plot
Coefficient of variation: Population Mean vs. Median Sample
Histogram
Skewness Coefficient
Chapter Review Chapter Review ®
CV = 100 ×Kurtosis
σ Box Plot
_
μ CoefficientCV = 100 × ¯
s
__
x
Histogram
Each chapter has a list of Standardized variable: Population
1. What are descriptive statistics? How do they differ from
Box Plot Sample
7. List strengths and weaknesses of each measure of center and
questions for student self- visual displays of data? xi − μ
zi = _____ zi = i
−¯
give its_____
x Excel
x function (if any): (a) midrange, (b) geometric
mean, ands (c) 10 percent trimmed mean.
review or for discussion. 2. Explain each concept: (a) center, (b) variability, andσ (c) shape.
8. (a) What is variability? (b) Name five measures of variability. List
3. (a) Why is sorting usually the first step in data analysis?
Q1 + Q3
_______
(b) Why is it useful to Midhinge:
begin a data analysis Midhinge = about
by thinking 2
the main characteristics (strengths, weaknesses) of each measure.
how the data were collected? 9. (a) Which standard deviation formula (population, sample) is
doa16287_ch04_100-157.indd 147
4. List strengths and weaknesses of each measure of center
n used most often? Why? (b) When is the coefficient of07:59
07/29/20 varia-
AM
∑ (xiand−¯ x )(y − ¯
y)
write its Excel function: (a) mean, (b) median, and (c)
i = 1mode.
_________________________ tion useful? ____
sXY
Sample correlation coefficient: r = _________ _________ or r =
√ √
n n sX sof
doa16287_ch04_100-157.indd 147
5. (a) Why must the deviations around the mean sum ∑ (to xi − x ) 2 ∑10.
¯
zero? (yi −(a)
y )To
¯ 2 what kind Y data does Chebyshev’s Theorem
07/29/20 apply?
07:59 AM
=1
(b) What is the position of the median in the data iarray when i =1 (b) To what kind of data does the Empirical Rule apply?
n is even? When n is odd? (c) Why is the mode of little use (c) What is an outlier? An unusual data value?
k k
in continuous data? (d) For
Weighted what type of ¯
mean: = ∑iswthe
xdata where ∑11.
j xj mode wj =(a) In a normal distribution, approximately what percent of
1.00
j =1 j =1
most useful? observations are within 1, 2, and 3 standard deviations of the
6. (a) What is a bimodal distribution? (b) Explain ktwo mean? (b) In a sample of 10,000 observations, about how
fj mways to
detect skewness. Grouped mean: x = ∑ ____
¯ j
many observations would you expect beyond 3 standard devi-
j =1 n ations of the mean?

x Chapter Review ®

1. What are descriptive statistics? How do they differ from 7. List strengths and weaknesses of each measure of center and
visual displays of data? give its Excel function (if any): (a) midrange, (b) geometric
2. Explain each concept: (a) center, (b) variability, and (c) shape. mean, and (c) 10 percent trimmed mean.
doa16287_ch04_100-157.indd 148
3. (a) Why is sorting usually the first step in data analysis? 8. (a) What is variability? (b) Name five measures of variability.
07/29/20 07:59List
AM
 12. (a) Write the mathematical formula for a standardized vari- 17. What does a correlation coefficient measure? What is its
able. (b) Write the Excel formula for standardizing a data range? Why is a correlation coefficient easier to interpret than
value in cell F17 from an array with mean Mu and standard a covariance?
deviation Sigma. 18. (a) Why is some accuracy lost when we estimate the mean

Student Learning?
13. (a) Why is it dangerous to delete an outlier? (b) When might
it be acceptable to delete an outlier?
14. (a) Explain how quartiles can measure both center and vari-
ability. (b) Why don’t we calculate percentiles for small
samples?
15. (a) Explain the method of medians for calculating quartiles.
(b) Write the Excel formula for the first quartile of an array
named XData.
16. (a) What is a box plot? What does it tell us? (b) What is the
role of fences in a box plot? (c) Define the midhinge and
interquartile range.
or standard deviation from grouped data? (b) Why do open-
ended classes in a frequency distribution make it impossible
to estimate the mean and standard deviation? (c) When would
grouped data be presented instead of the entire sample of
raw data?
19. (a) What is the skewness coefficient of a normal distribution?
A uniform distribution? (b) Why do we need a table for
sample skewness coefficients that is based on sample size?
20. (a) What is kurtosis? (b) Sketch a platykurtic population, a
leptokurtic population, and a mesokurtic population. (c) Why
can’t we rely on a histogram to assess kurtosis?

Chapter Exercises Chapter Exercises ®

Exercises give students an oppor-

4.54 (a) For each data set, calculate the mean, median, and mode. Find Bob’s standardized z-score. (b) By the Empirical Rule,
tunity to test their understanding (b) Which, if any, of these three measures is the weakest is Bob’s SAT score unusual?
of the chapter material. Exercises indicator of a “typical” data value? Why? 4.61 Find the data value that corresponds to each of the follow-
ing z-scores.
a. Number of e-mail accounts (12 students): 1, 1, 1, 1, 2, 2,
are included at the ends of sec- 2, 3, 3, 3, 3, 3 a. Final exam scores: Allison’s z-score = 2.30, μ = 74, σ = 7
tions and at the ends of chapters. b. Number of siblings (5 students): 0, 1, 2, 2, 10
c. Asset turnover ratio (8 retail firms): 1.85, 1.87, 2.02,
b. Weekly grocery bill: James’ z-score = −1.45, μ = $53,
σ = $12
Some exercises contain data sets, 2.05, 2.11, 2.18, 2.29, 3.01 c. Daily video game play time: Eric’s z-score = −0.79,
μ = 4.00 hours, σ = 1.15 hours
identified by data set icons. Data 4.55 If the mean asset turnover for retail firms is 2.02 with a
standard deviation of 0.22, without assuming a normal 4.62 The average time a Boulder High varsity lacrosse player
First Pages
sets can be accessed through distribution, within what range will at least 75% of retail plays in a game is 30 minutes with a standard deviation First Pages
of
firms’ asset turnover fall? 7 minutes. Nolan’s playing time in last week’s game against
Connect and used to solve prob- 4.56 For each data set: (a) Find the mean, median, and mode. Fairview was 48 minutes. (a) Calculate the z-score for
(b) Which, if any, of these three measures is the weakest Nolan’s playing time against Fairview. (b) By the Empirical
lems in the text. indicator of a “typical” data value? Why? Rule, was Nolan’s playing time unusual when compared to
156 Applied Statistics in Business and Economics the typical playing time?
a. 100 m dash times (n = 6 top runners): 9.87, 9.98, 10.02,
154 Applied
10.15, Statistics in Business and Economics
10.36, 10.36 4.63 The number of blueberries in a blueberry muffin baked by
Once the data are imported to VehicleData, we can view the data, Be careful of the syntax,
EarthHarvest e.g., the
Bakeries cansquare
range brackets ]. You
from 18[ to can also refer
30 blueberries.
b. Number of children (n = 13 families): 0, 1, 1, 2, 2, 2, 2,
calculate statistics, and so on. For example, to calculate average to columns
(a) Usein the
yourEmpirical
data frame by number,
Rule separated
to estimate by commas
the standard (or
devia-
Related 2, 2, Reading
2, 2, 2, 6
city miles per gallon for our 50 vehicles (the column named by a colon : instead
tion of of a comma
the number if your columns
of blueberries in a are contiguous).
muffin. (b) What
c. Number of cars in driveway (n
the=function
8 homes): 0, 0,and
1, 1, 2,
Doane, inDavid
CityMPG the imported
P., and data),
Lori weE. use
Seward. mean()
“Measuring name
Skewness: assumption
Pukelsheim, did you“The
Friedrich. makeThree
> summary(VehicleData[c(6:8)])
about Sigma
the distribution
Rule.” Theof the num-
American
theForgotten
column2, 3,in5the data frame followed by $ (the dollar symbol) as ber of blueberries?
A Statistic?” Journal of Statistics Education 19, no. 2 (2011). Statistician
HP 48, no. 2 (May 1994), pp. 88–91.
Engine CityMPG
4.57
follows: During a rock concert, the noise level (in decibels) in front
Freund, Unless otherwise instructed,Introductory
you may use any desired statistical
rowJohn
seats E.,
has and Benjamin M. with
Perles. “A Newdeviation
Look at Verzani,
Note: John. Using 2nd ed.
a mean of 95 dB a standard Min. software
:106.0 Min. Randfor
for calculations :1400in the following
graphs Min. Statistics.
:13.00
problems.
> mean(VehicleData$CityMPG)
Quartilesof 8 of
dB.Ungrouped Data.” The
Without assuming American
a normal Statistician
distribution, 41,the
find no. Chapman-Hall,
1st Qu. :182.5 2014. 1st Qu. :2000 1st Qu. :18.00
3[1] (August
21.98
minimum1987),percentage
pp. 200–203.of noise level readings within 3 stan- Median :249.0 Median :2500 Median :22.00

More Learning mean.names (as they do here), we can DESCRIBING Mean :262.3 DATA Mean :2898 Mean :21.98
If ourdard datadeviations
frame columnsof the have
3rd Qu. :308.0 3rd Qu. :3500 3rd Qu. :26.00
4.58 Bags of statistics
getCHAPTER
summary jelly beans forhave a mean
variables weight
interest of
of Resources 396 the
using gmsummary()
with a 4.64 Below are monthly rents:6200
paid byMax.30 students who live® off
4 More Learning
Resources
Max. :602.0 Max. :31.00
standard deviation of 5 gm. Use Chebyshev’s Theorem to campus. (a) ®Find the mean,
command.
You can For example:
access these LearningStats demonstrations through McGraw-Hill’s Connect to help you median, and mode. (b) Do the
find a lower bound for the number of bags in a sample of You can create of
measures a graph
centraland export itagree?
tendency from the Plots tab
Explain. (c)(lower
Calculateright
understand descriptive statistics.
> summary(VehicleData[c(''Weight'',''Length'')])
200 that weigh between 386 and 406 gm. pane the
in R)standard
to paste it into your(d) written
Sort report. For example, the we can
LearningStats provides a Based on experience,
4.59 Weight the Ball Corporation’s aluminum can
Length create(e)
a simple
deviation.
boxoutliers
Are there plot andorhistogram
and standardize
withvalues?
unusual data optional(f)labels
Using
data.
forthe
the
Topic LearningStats Demonstrations
means for Connect users to Min. manufacturing
:2385
1st Qu.that :3356
the metal
facility :151.1
Min.
thickness
1st
OverviewQu.
in Ft. Atkinson, Wisconsin, knows
of incoming shipments has a mean
:181.9
axes and graph titles:
Empirical Rule, do you think the data could be from a nor-
mal population?
Describing Data Rents
explore data and concepts at Medianof 0.2731
:3662 mmMedian :192.2 deviation of 0.000959 mm.
with a standard > boxplot(VehicleData$Weight, ylab=''Pounds'', main=''Vehicle Weight'')
Using MegaStat
> hist(VehicleData$Weight, xlab=''Pounds'', main=''Vehicle
Mean (a) A:3954 certain Mean
shipment:190.9
has a diameter of 0.2761. Find the Using 730
Minitab 730 730 930 700 570 Weight'')
their own pace. Applications 3rd Qu.standardized
:4661 3rd Qu. for:198.7
z-score this shipment. (b) Is this an outlier? Using 690
R 1,030 740 620 720 670
Max. SAT:5917 Max. :231.9 class of 2010 at Oxnard Uni-
that relate to the material in the 4.60 scores for the entering
Descriptive statistics
versity were normally distributed with a mean of 1340 and
560
Basic Statistic
600 620
740 650
760
660
690
850
710
930
500
chapter are identified by topic a standard deviation of 90. Bob’s SAT score was 1430. (a) Quartiles
730 800 820 840 720 700
Box Plot Simulation
at the end of each chapter. Grouped Data
Significant Digits
ScreenCam Tutorials Using MegaStat
Excel Descriptive Statistics
Excel Scatter Plots
doa16287_ch04_100-157.indd 149 07/29/20 07:59 AM
Key: = PowerPoint = Excel = PDF = ScreenCam Tutorials

Exam Review Questions
At the end of a group of chapters,
students can review the material
they covered in those chapters.
This provides them with an oppor-
tunity to test themselves on their
grasp of the material.
Software Supplement
Descriptive Statistics Using Megastat On the new menu, enter the data range (in this case C4:C37) in the Input
You can obtain descriptive statistics (and more) from MegaStat, as range field (or highlight the data block on the worksheet). MegaStat
illustrated in Figure 4.33. Click the Add-Ins tab on the top menu, and
Exam
then click Review Questions
on the MegaStat for
icon (left side Chapters
of the top menu in1–4 this
offers you various statistics and visual displays, including a dot plot
and stem-and-leaf. Compare Excel and MegaStat to see similarities
example). On the list of MegaStat procedures, click Descriptive Statistics.
1. Which type of statistic (descriptive, inferential) is each of the
and differences in their interfaces and results.
4. Which data type (categorical, numerical) is each of the following?
Figure 4.33
following? a. Your current credit card balance.
a. Estimating the default rate on all U.S. mortgages from a b. Your college major.
MegaStat’s
randomDescriptive
sample of 500 loans. c. Your car’s odometer mileage reading today.
Statistics JDPower
b. Reporting the percent of students in your statistics class
who use Verizon.
5. Give the type of measurement (nominal, ordinal, interval,
ratio) for each variable.
c. Using a sample of 50 iPhones to predict the average battery
a. Length of time required for a randomly chosen vehicle to
life in typical usage.
2. Which is not an ethical obligation of a statistician? Explain.
cross a toll bridge.
b. Student’s ranking of five cell phone service providers.
a. To know and follow accepted procedures. c. The type of charge card used by a customer (Visa, Mastercard,
b. To ensure data integrity and accurate calculations. AmEx, Other).
c. To support client wishes in drawing conclusions from 6. Tell if each variable is continuous or discrete.
the data.
a. Tonnage carried by an oil tanker at sea.
3. “Driving without a seat belt is not risky. I’ve done it for 25 b. Wind velocity at 7 o’clock this morning.
years without an accident.” This best illustrates which fallacy? c. Number of text messages you received yesterday.
a. Unconscious bias. 7. To choose a sample of 12 students from a statistics class of
Source: MegaStat
b. Conclusion from a small sample. 36 students, which type of sample (simple random, systematic,
c. Post hoc reasoning. cluster, convenience) is each of these?

xi

doa16287_ch04_100-157.indd 156 07/29/20 07:59 AM

doa16287_ch04_100-157.indd 154 07/29/20 07:59 AM

Instructors: Student Success Starts with You
Tools to enhance your unique voice
Want to build your own course? No problem. Prefer to use our
turnkey, prebuilt course? Easy. Want to make changes throughout the
65%
Less Time
semester? Sure. And you’ll save time with Connect’s auto-grading too.
Grading

Study made personal

Incorporate adaptive study resources like
SmartBook® 2.0 into your course and help your
students be better prepared in less time. Learn
more about the powerful personalized learning
experience available in SmartBook 2.0 at
www.mheducation.com/highered/connect/smartbook

Laptop: McGraw Hill; Woman/dog: George Doyle/Getty Images

Affordable solutions, Solutions for

added value your challenges
Make technology work for you with A product isn’t a solution. Real
LMS integration for single sign-on access, solutions are affordable, reliable,
mobile access to the digital textbook, and come with training and
and reports to quickly show you how ongoing support when you need
each of your students is doing. And with it and how you want it. Visit www.
our Inclusive Access program you can supportateverystep.com for videos
provide all these tools at a discount to and resources both you and your
your students. Ask your McGraw Hill students can use throughout the
representative for more information. semester.

Padlock: Jobalou/Getty Images Checkmark: Jobalou/Getty Images

Students: Get Learning that Fits You
Effective tools for efficient studying
Connect is designed to make you more productive with simple, flexible, intuitive tools that maximize
your study time and meet your individual learning needs. Get learning that works for you with Connect.

Study anytime, anywhere “I really liked this

Download the free ReadAnywhere app and access your app—it made it easy
online eBook or SmartBook 2.0 assignments when it’s to study when you
convenient, even if you’re offline. And since the app don't have your text-
automatically syncs with your eBook and SmartBook 2.0
assignments in Connect, all of your work is available book in front of you.”
every time you open it. Find out more at
www.mheducation.com/readanywhere - Jordan Cunningham,
Eastern Washington University

Everything you need in one place

Your Connect course has everything you need—whether reading on
your digital eBook or completing assignments for class, Connect makes
it easy to get your work done.

Calendar: owattaphotos/Getty Images

Learning for everyone

McGraw Hill works directly with Accessibility Services
Departments and faculty to meet the learning needs
of all students. Please contact your Accessibility
Services Office and ask them to email
accessibility@mheducation.com, or visit
www.mheducation.com/about/accessibility
for more information.
Top: Jenner Images/Getty Images, Left: Hero Images/Getty Images, Right: Hero Images/Getty Images
Additional Connect Features
Excel Data Sets A convenient feature is the inclusion of an Excel data file link in many problems using data files in their
calculation. The link allows students to easily launch into Excel, work the problem, and return to Connect to key in the
answer.

Guided Examples These narrated video walkthroughs provide students with step-by-step guidelines for solving selected
exercises similar to those contained in the text. The student is given personalized instruction on how to solve a problem by
applying the concepts presented in the chapter. The narrated voiceover shows the steps to take to work through an exercise.
Students can go through each example multiple times if needed.

xiv
What Resources are Available for Students?
The following software tools are available to assist students in understanding concepts and solving problems.

LearningStats
LearningStats allows students to explore data and con-
cepts at their own pace. It includes demonstrations,
simulations, and tutorials that can be downloaded from
Connect.

MegaStat® for Excel®

Access Card (ISBN: 0077426274) or online purchase at www.mhhe.com/megastat.
MegaStat is a full-featured Excel add-in that is available with this text. It performs statistical analyses within an Excel work-
book. It does basic functions such as descriptive statistics, frequency distributions, and probability calculations as well as
hypothesis testing, ANOVA, and regression.
MegaStat output is carefully formatted, and ease-of-use features include Auto Expand for quick data selection and Auto
Label detect. Because MegaStat is easy to use, students can focus on learning statistics without being distracted by the
software. MegaStat is always available from Excel’s main menu. Selecting a menu item pops up a dialog box. MegaStat is
updated continuously to work with the latest versions of Excel for Windows and Macintosh users.

Minitab® and Minitab Express®

Free trials and academic versions are available from Minitab at http://www.minitab.com/en-us/academic/.

R and RStudio
A sophisticated programming language for statistical computing and graphics plus an integrated development environment.
This textbook offers detailed instructions for downloading, installing, and using free versions of R (https://www.r-project.
org/) and RStudio (https://rstudio.com/).

xv
What Resources are Available for Instructors?
Instructor resources are available through the Connect course at connect.mheducation.com. Resources include a complete
Instructor’s Manual in Word format, the complete Test Bank. Instructor PowerPoint slides, text art files, and more.

Tegrity: Lectures 24/7

Tegrity in Connect is a tool that makes class time available 24/7 by automatically capturing every lecture. With a simple
one-click start-and-stop process, you capture all computer screens and corresponding audio in a format that is easy to
search, frame by frame. Students can replay any part of any class with easy-to-use, browser-based viewing on a PC, Mac,
iPod, or other mobile device.
Educators know that the more students can see, hear, and experience class resources, the better they learn. In fact, studies
prove it. Tegrity’s unique search feature helps students efficiently find what they need, when they need it, across an entire
semester of class recordings. Help turn your students’ study time into learning moments immediately supported by your lec-
ture. With Tegrity, you also increase intent listening and class participation by easing students’ concerns about note-taking.
Using Tegrity in Connect will make it more likely you will see students’ faces, not the tops of their heads.

Test Builder in Connect

Available within Connect, Test Builder is a cloud-based tool that enables instructors to format tests that can be printed or
administered within a LMS. Test Builder offers a modern, streamlined interface for easy content configuration that matches
course needs, without requiring a download.
Test Builder allows you to:
∙ access all test bank content from a particular title.
∙ easily pinpoint the most relevant content through robust filtering options.
∙ manipulate the order of questions or scramble questions and/or answers.
∙ pin questions to a specific location within a test.
∙ determine your preferred treatment of algorithmic questions.
∙ choose the layout and spacing.
∙ add instructions and configure default settings.
Test Builder provides a secure interface for better protection of content and allows for just-in-time updates to flow
directly into assessments.

Remote Proctoring & Browser-Locking Capabilities

New remote proctoring and browser-locking capabilities, hosted by Proctorio within Connect, provide control of the assess-
ment environment by enabling security options and verifying the identity of the student.
Seamlessly integrated within Connect, these services allow instructors to control students’ assessment experience by
restricting browser activity, recording students’ activity, and verifying students are doing their own work.
Instant and detailed reporting gives instructors an at-a-glance view of potential academic integrity concerns, thereby
avoiding personal bias and supporting evidence-based claims.

McGraw Hill Customer Experience Information

For Customer Support, call 800-331-5094 or visit www.mhhe.com/support. One of our customer experience team members
will be able to assist you in a timely fashion.

xvi
 Rev.Confirming Pages

Acknowledgments
The authors would like to acknowledge some of the many people who have helped with this book. Thomas W. Lauer and
Floyd G. Willoughby permitted quotation of a case study. Morgan Elliott, Karl Majeske, Robin McCutcheon, Kevin Mur-
phy, John Sase, T. J. Wharton, and Kenneth M. York permitted questionnaires to be administered in their classes. Mark
Isken, Ron Tracy, and Robert Kushler gave generously of their time as expert statistical consultants. Jonathan G. Koomey of
E.O. Lawrence Berkeley National Laboratory offered valuable suggestions on visual data presentation.
We are grateful to Farrukh Abbas for his careful scrutiny of the text and for offering ideas on improving the text and
exercises. Mark Isken has reliably provided Excel expertise and has suggested health care applications for examples and
case studies. John Savio and the Michigan State Employees Credit Union provided ATM data. The Siena Research Institute
has made its poll results available. J.D. Power and Associates generously provided permission to use vehicle quality data.
The Public Interest Research Group in Michigan (PIRGIM) has generously shared data from its field survey of prescription
drug prices.
Phil Rogers has offered numerous suggestions for improvement in both the textbook exercises and Connect. Milo A.
Schield shared his research on “quick rules” for measuring skewness from summarized data. We owe special thanks to
Aaron Kennedy and Dave Boennighausen of Noodles & Company; to Mark Gasta, Anja Wallace, and Clifton Pacaro of
Vail Resorts; to Jim Curtin and Gordon Backman of Ball Corporation; and to Santosh Lakhan from The Verdeo Group for
providing suggestions and access to data for Mini Cases and examples. For reviewing the material on quality, we wish to
thank Kay Beauregard, administrative director at William Beaumont Hospital, and Ellen Barnes and Karry Roberts of Ford
Motor Company. Amy Sheikh provided a new Facebook Friends data set, along with other excellent suggestions and reports
from the “front lines” of her classes.
A special debt of gratitude is due to Noelle Bathurst, Harper Christopher, Amy Gehl, and Ryan McAndrews for their
direction and support and Harvey Yep and Jamie Koch for managing the text and Connect pieces of the project. Thanks
to the many reviewers who provided such valuable feedback including criticism that made the book better, some of whom
reviewed several previous editions of the text. Any remaining errors or omissions are the authors’ responsibility. Thanks
too, to the participants in our focus groups and symposia on teaching business statistics, who have provided teaching ideas
and insights from their experiences with students in diverse contexts. We hope you will be able to see in our book and the
teaching package consideration of those ideas and insights.

Farrukh Abbas, National University of Modern Languages
(NUML), Islamabad, Pakistan
Heather Adams, University of Colorado—Boulder
Sung Ahn, Washington State University
Mostafa Aminzadeh, Towson University
Scott Bailey, Troy University
Hope Baker, Kennesaw State University
Saad Taha Bakir, Alabama State University
Steven Bednar, Elon University
Adam Bohr, University of Colorado—Boulder
Katherine Broneck, Pima Community College—Downtown
Alan Cannon, University of Texas—Arlington
Deborah Carter, Coahoma Community College
Kevin Caskey, SUNY—New Paltz
Michael Cervetti, University of Memphis
Paven Chennamaneni, University of Wisconsin—Whitewater
Alan Chesen, Wright State University
Wen-Chyuan Chiang, University of Tulsa
Chia-Shin Chung, Cleveland State University
Joseph Coleman, Wright State University—Dayton
Robert Cutshall, Texas A&M University—Corpus Christi
Terry Dalton, University of Denver
Douglas Dotterweich, East Tennessee State University
Jerry Dunn, Southwestern Oklahoma State University
Michael Easley, University of New Orleans
Jerry Engeholm, University of South Carolina—Aiken
Mark Farber, University of Miami
Soheila Kahkashan Fardanesh, Towson University
Mark Ferris, St. Louis University
Stergios Fotopoulos, Washington State University
Vickie Fry, Westmoreland County Community College
Joseph Fuhr, Widener University
Bob Gillette, University of Kentucky
Malcolm Gold, Avila University
Don Gren, Salt Lake City Community College
Karina Hauser, University of Colorado—Boulder
Eric Hernandez, Miami Dade College
Clifford Hawley, West Virginia University
Yijun He, Washington State University
Natalie Hegwood, Sam Houston State University

xvii

doa16287_fm_i-1.indd xvii 05/27/21 06:06 PM

Allen Humbolt, University of Tulsa
Patricia Igo, Northeastern University
Alam M. Imam, University of Northern Iowa
Marc Isaacson, Augsburg College
Kishen Iyengar, University of Colorado—Boulder
Christopher Johnson, University of North Florida
Jennifer Johnson, San Jose State
Ronald Johnson, Central Alabama Community College
Linda Jones, Maryville University
Jerzy Kamburowski, University of Toledo
Bob Kitahara, Troy University
Drew Koch, James Madison University
Mohammad Kazemi, University of North Carolina–Charlotte
Agnieszka Kwapisz, Montana State University
Mary Beth Camp, Indiana University—Bloomington
Kenneth Lawrence, New Jersey Institute of Technology
Bob Lynch, University of Northern Colorado
Bradley McDonald, Northern Illinois University
Richard McGowan, Boston College
Kelly McKillop, University of Massachusetts
Larry McRae, Appalachian State University
Robert Mee, University of Tennessee—Knoxville
Mahour Mellat Parast, University of North Carolina–Pembroke
John Miller, Sam Houston State University
Shelly Moore, College of Western Idaho
James E. Moran Jr., Oregon State University
Geraldine Moultine, Northwood University
Gourab Mukherjee, University of Southern California
Adam Munson, University of Florida
Joshua Naranjo, Western Michigan University
Anthony Narsing, Macon State College
Robert Nauss, University of Missouri–St. Louis
Pin Ng, Northern Arizona University
Thomas Obremski, University of Denver
Rahmat Ola Tavallai, Walsh University
Grace Onodipe, Georgia Gwinnett College
Ceyhun Ozgur, Valparaiso University
Ed Pappanastos, Troy University
Nitin Paranjpe, Oakland University
Eddy Patuwo, Kent State University
xviii
John Pickett, University of Arkansas—Little Rock
James Pokorski, Virginia Polytechnic Institute &
State University
Stephan Pollard, California State University—Los Angeles
Claudia Pragman, Minnesota State University
Tammy Prater, Alabama State University
Michael Racer, University of Memphis
Azar Raiszadeh, Chattanooga State Community College
Mohammad Reza Oskoorouchi, California State University—San
Marcos
Phil Rogers, University of Southern California
Milo A. Schield, Augsburg College
Sue Schou, Idaho State University
Elizabeth Scofidio, Colorado State University
Sankara N. Sethuraman, Augusta State University
Don Sexton, Columbia University
Thomas R. Sexton, Stony Brook University
Murali Shanker, Kent State University
Gary W. Smith, Florida State University
Courtenay Stone, Ball State University
Paul Swanson, Illinois Central College
Bedassa Tadesse, University of Minnesota–Duluth
Rahmat Tavallali, Walsh University
Deborah Tesch, Xavier University
Dharma S. Thiruvaiyaru, Augusta State University
Frank Tian Xie, University of South Carolina–Aiken
Bhavneet Walia, Western Illinois University
Michael Urizzo, New Jersey City University
Jesus M. Valencia, Slippery Rock University
Rachel Webb, Portland State University
Simone A. Wegge, City University of New York
Chao Wen, Eastern Illinois University
Alan Wheeler, University of Missouri—St. Louis
Blake Whitten, University of Iowa
Charles Wilf, Duquesne University
Anne Williams, Gateway Community College
Janet Wolcutt, Wichita State University
Frank Xie, University of South Carolina—Aiken
Ye Zhang, Indiana University–Purdue University Indianapolis
Mustafa R. Yilmaz, Northeastern University
Another random document with
no related content on Scribd:
 VEAL CUTLETS, OR COLLOPS, À LA FRANÇAISE. (ENTRÉE.)

Cut the veal into small, thin, round collops of equal size, arrange
them evenly in a sauté-pan, or in a small frying-pan, and sprinkle a
little fine salt, white pepper, and grated nutmeg on them. Clarify, or
merely dissolve in a clean saucepan with a gentle degree of heat, an
ounce or two of good butter, and pour it equally over the meat. Set
the pan aside until the dinner-hour, then fry the collops over a clear
fire, and when they are lightly browned, which will be in from four to
five minutes, lift them into a hot dish, and sauce them with a little
Espagnole, or with a gravy made quickly in the pan, and flavoured
with lemon-juice and cayenne. They are excellent even without any
sauce.
3 to 4 minutes.
 SCOTCH COLLOPS. (ENTRÉE.)

Prepare the veal as for the preceding receipt, but dip the collops
into beaten egg and seasoned bread-crumbs, and fry them directly in
good butter, over a moderate fire, of a light golden brown; drain them
well in lifting them from the pan, and sauce them like the collops à la
Française.
 VEAL CUTLETS À LA MODE DE LONDRES, OR, LONDON
FASHION. (ENTRÉE.)

Raise the flesh entire from the upper side of the best end of a neck
of veal, free it from the skin, and from the greater portion of the fat,
slice it equally into cutlets little more than a quarter of an inch thick,
brush them with egg, strew them with fine bread-crumbs, and fry
them of a light brown. Toast, or fry apart as many small slices of
bacon as there are cutlets, and let them be trimmed nearly to the
same shape; place them alternately on their edges round the inside
of a hot dish (so as to form a sort of chain), and pour into the middle
some rich gravy made in the pan, and very slightly flavoured with
eschalot; or substitute for this some good brown mushroom sauce.
Savoury herbs, grated lemon-rind, nutmeg or mace, salt, and white
pepper or cayenne, should be mixed with the bread-crumbs, in the
proportions directed at page 213, for cutlets of calf’s head; or they
may be varied at pleasure. A cheek of bacon is best adapted to this
dish.
 SWEETBREADS SIMPLY DRESSED. (ENTRÉE.)

In whatever way sweetbreads are dressed, they should first be

well soaked in lukewarm water, then thrown into boiling water to
blanch them, as it is called, and to render them firm. If lifted out after
they have boiled from five to ten minutes according to their size, and
laid immediately into fresh spring water to cool, their colour will be
the better preserved. They may then be gently stewed for three
quarters of an hour in veal gravy, which with the usual additions of
cream, lemon, and egg-yolks, may be converted into a fricassee
sauce for them when they are done; or they may be lifted from it,
glazed, and served with good Spanish gravy; or, the glazing being
omitted, they may be sauced with the sharp Maître d’Hôtel sauce of
page 117. They may also be simply floured, and roasted in a Dutch
oven, being often basted with butter, and frequently turned. A full
sized sweetbread, after having been blanched, will require quite
three quarters of an hour to dress it.
Blanched 5 to 10 minutes. Stewed 3/4 hour or more.
 SWEETBREAD CUTLETS. (ENTRÉE.)

Boil the sweetbreads for half an hour in water or veal broth, and
when they are perfectly cold, cut them into slices of equal thickness,
brush them with yolk of egg, and dip them into very fine bread-
crumbs seasoned with salt, cayenne, grated lemon-rind, and mace;
fry them in butter of a fine light brown, arrange them in a dish placing
them high in the centre, and pour under them a gravy made in the
pan, thickened with mushroom powder and flavoured with lemon-
juice; or, in lieu of this, sauce them with some rich brown gravy, to
which a glass of sherry or Madeira has been added. When it can be
done conveniently, take as many slices of a cold boiled tongue as
there are sweetbread cutlets; pare the rind from them, trim them into
good shape, and dress them with the sweetbreads, after they have
been egged and seasoned in the same way; and place each cutlet
upon a slice of tongue when they are dished. For variety, substitute
croutons of fried bread stamped out to the size of the cutlets with a
round or fluted paste or cake cutter. The crumb of a stale loaf, very
evenly sliced, is best for the purpose.
 STEWED CALF’S FEET.

(Cheap and Good.)

This is an excellent family dish, highly nutritious, and often very
inexpensive, as the feet during the summer are usually sold at a low
rate. Wash them with nicety, divide them at the joint, and split the
claws; arrange them closely in a thick stewpan or saucepan, and
pour in as much cold water as will cover them about half an inch:
three pints will be sufficient for a couple of large feet. When broth or
stock is at hand, it is good economy to substitute it for the water, as
by this means a portion of strong and well-flavoured jellied gravy will
be obtained for general use, the full quantity not being needed as
sauce for the feet. The whole preparation will be much improved by
laying a thick slice of the lean of an unboiled ham, knuckle of bacon,
hung beef, or the end of a dried tongue, at the bottom of the pan,
before the other ingredients are added; or, when none of these are at
hand, by supplying the deficiency with a few bits of lean beef or veal:
the feet being of themselves insipid, will be much more palatable
with one or the other of these additions. Throw in from half to three
quarters of a teaspoonful of salt when they begin to boil, and after
the scum has been all cleared off, add a few branches of parsley, a
little celery, one small onion or more, stuck with half a dozen cloves,
a carrot or two, a large blade of mace, and twenty corns of whole
pepper; stew them softly until the flesh will part entirely from the
bones; take it from them, strain part of the gravy, and skim off all the
fat, flavour it with catsup or any other store sauce, and thicken it,
when it boils, with arrow-root or flour and butter; put in the flesh of
the feet, and serve the dish as soon as the whole is very hot. A glass
of wine, a little lemon juice, and a few forcemeat balls, will convert
this into a very superior stew; a handful of mushroom-buttons also
simmered in it for half an hour before it is dished, will vary it
agreeably.
Calf’s feet (large), 2; water, 3 pints; salt, 1/2 to 1/3 teaspoonful;
onions, 1 to 3; cloves, 6; peppercorns, 20; mace, large blade; little
celery and parsley; carrots, 1 or 2: stewed softly, 2-1/2 to 3-1/4
hours. Mushroom catsup, 1 tablespoonful; flour, or arrow-root, 1
large teaspoonful; butter, 1 to 2 oz. Cayenne, to taste.
 CALF’S LIVER STOVED, OR STEWED.

From three to four pounds of the best part of the liver will be
sufficient for a dish of moderate size. First lard it quite through by the
directions of page 181, with large lardoons, rolled in a seasoning of
spice, and of savoury herbs very finely minced; then lay it into a
stewpan or saucepan just fitted to its size, and pour in about half a
pint of broth or gravy; heat it very gently, and throw in, when it begins
to simmer, a sliced carrot, a small onion cut in two, a small bunch of
parsley, and a blade of mace; stew the liver as softly as possible
over a very slow fire from two hours and a half to three hours;
thicken the gravy with a little brown roux (see page 107), or with a
dessertspoonful of browned flour; add a couple of glasses of white
wine, and a little spice if needed, and serve it very hot, after having
taken out the herbs and vegetable.
The liver may be stewed without being larded; it may likewise be
browned all over in a carefully made roux, before the gravy is poured
to it: this must then be made to boil, and be added in small portions,
the stewpan being well shaken round as each is thrown in. The wine
can be altogether omitted; or a wineglassful of port mixed with a little
lemon-juice, may take the place of sherry. After the liver has been
wiped very dry, minced herbs may be strewed thickly over it before it
is laid into the stewpan; and it may be served in its own gravy, or
with a sauce piquante.
Liver, 3 to 4 lbs: 2 to 3 hours.
 TO ROAST CALF’S LIVER.

Take the whole or part of a fine white sound liver, and either lard it
as a fricandeau upon the surface, or with large strips of highly-
seasoned bacon in the inside (see Larding, page 181); or should
either of these modes be objected to, merely wrap it in a well
buttered paper, and roast it from an hour to an hour and a quarter at
a moderate distance from a clear fire, keeping it constantly basted.
Remove the paper, and froth the liver well from ten to fifteen minutes
before it is done. It should be served with a sauce of some piquancy,
such as a poivrade, or brown eschalot, in addition to some good
gravy. French cooks steep the liver over-night in vinegar, with a
sliced onion and branches of savoury herbs laid over it: this whitens
and renders it firm. As an economical mode, some small bits of the
liver may be trimmed off, floured, and lightly fried with a sliced onion,
and stewed down for gravy in three quarters of a pint of water which
has been poured into the pan, with the addition of a few
peppercorns, and a small bunch of herbs. A seasoning of salt must
not be forgotten, and a little lemon pickle, or juice, would generally
be considered an improvement.
1 to 1-1/4 hour.
 BLANQUETTE OF VEAL OR LAMB, WITH MUSHROOMS.

(ENTRÉE.)
Slice very thin the white part of some cold veal, divide and trim it
into scallops not larger than a shilling, and lay it into a clean
saucepan or stewpan. Wipe with a bit of new flannel and a few
grains of salt, from a quarter to half a pint of mushroom-buttons, and
slice them into a little butter which just begins to simmer; stew them
in it from twelve to fifteen minutes, without allowing them to take the
slightest colour; then lift them out and lay them on the veal. Pour
boiling to them a pint of sauce tournée (see page 108); let the
blanquette remain near, but not close to the fire for awhile: bring it
nearer, heat it slowly, and when it is on the point of boiling mix a
spoonful or two of the sauce from it with the well beaten yolks of four
fresh eggs; stir them to the remainder; add the strained juice of half a
small lemon; shake the saucepan above the fire until the sauce is
just set, and serve the blanquette instantly.
Cold veal, 3/4 lb.; mushrooms, 1/4 to 1/2 pint: stewed in 1-1/2 oz.
butter, 12 to 15 minutes. Sauce tournée, or thickened veal gravy, 1
pint; yolks of eggs, 4; lemon-juice, 1 tablespoonful.
Obs.—Any white meat may be served en blanquette. The
mushrooms are not indispensable for it, but they are always a great
improvement. White sauce substituted for the thickened veal gravy
will at once convert this dish into an inexpensive English fricassee.
Mace, salt, and cayenne, must be added to either preparation,
should it require seasoning.
 MINCED VEAL.

When there is neither gravy nor broth at hand, the bones and
trimmings of the meat must be boiled down to furnish what is
required for the mince. As cold meat is very light in weight, a pound
of the white part of the veal will be sufficient for a dish, and for this
quantity a pint of gravy will be needed. Break down the bones of the
joint well, add the trimmings of the meat, a small bunch of savoury
herbs, a slice or two of carrot or of celery, a blade of mace, a few
white peppercorns, and a bit or two of lean ham, boiled, or unboiled
if it can be had, as either will improve the flavour of the mince. Pour
to these a pint and a half of water, and stew them gently for a couple
of hours; then strain off the gravy, let it cool, and clear it entirely from
the fat. Cut the white part of the veal small with a very sharp knife,
after all the gristle and brown edges have been trimmed away. Some
persons like a portion of fat minced with it, others object to the
addition altogether. Thicken the gravy with a teaspoonful and a half
of flour smoothly mixed with a small slice of butter, season the veal
with a saltspoonful or more of salt, and half as much white pepper
and grated nutmeg, or pounded mace; add the lightly-grated rind of
half a small lemon; mix the whole well, put it into the gravy, and heat
it thoroughly by the side of the fire without allowing it to boil; serve it
with pale toasted sippets in and round the dish. A spoonful or two of
cream is always an improvement to this mince.
 MINCED VEAL AND OYSTERS.

The most elegant mode of preparing this dish is to mince about a

pound of the whitest part of the inside of a cold roast fillet or loin of
veal, to heat it without allowing it to boil, in a pint of rich white sauce,
or béchamel, and to mix with it at the moment of serving, three
dozens of small oysters ready bearded, and plumped in their own
strained liquor, which is also to be added to the mince; the requisite
quantity of salt, cayenne, and mace should be sprinkled over the
veal before it is put into the sauce. Garnish the dash with pale fried
sippets of bread, or with fleurons[77] of brioche, or of puff-paste.
Nearly half a pint of mushrooms minced, and stewed white in a little
butter, may be mixed with the veal instead of the oysters; or should
they be very small they may be added to it whole: from ten to twelve
minutes will be sufficient to make them tender. Balls of delicately
fried oyster-forcemeat laid round the dish will give another good
variety of it.
77. Fleurons, flowers, or flower-like figures, cut out with tin shapes.
Veal minced, 1 lb.; white sauce, 1 pint; oysters, 3 dozen, with their
liquor; or mushrooms, 1/2 pint, stewed in butter 10 to 12 minutes.
 VEAL-SYDNEY. (GOOD.)

Pour boiling on an ounce and a half of fine bread-crumbs nearly

half a pint of good veal stock or gravy, and let them stand till cool;
mix with them then, two ounces of beef-suet shred very small, half a
pound of cold roast veal carefully trimmed from the brown edges,
skin, and fat, and finely minced; the grated rind of half a lemon,
nearly a teaspoonful of salt, a little cayenne, the third of a
teaspoonful of mace or nutmeg, and four well-beaten eggs. Whisk up
the whole well together, put it into a buttered dish, and bake it from
three quarters of an hour to an hour. Cream may be used instead of
gravy when more convenient, but this last will give the better flavour.
A little clarified butter put into the dish before the other ingredients
are poured in will be an improvement.
Bread-crumbs, 1-1/2 oz.; gravy or cream, nearly 1/2 pint; beef-
suet, 2 oz.; cold veal, 1/2 lb.; rind of 1/2 lemon; salt, small
teaspoonful; third as much mace and nutmeg; little cayenne; eggs, 4
large or 5 small: 3/4 to 1 hour.
 FRICASSEED VEAL.

Divide into small, thick, handsome slices of equal size, about a

couple of pounds of veal, quite free from fat, bone, and skin; dissolve
a couple of ounces of butter in a wide stewpan, and just as it begins
to boil lay in the veal, and shake it over the fire until it is quite firm on
both sides, but do not allow it to take the slightest colour. Stir in a
tablespoonful of flour, and when it is well mixed with the cutlets, pour
gradually to them, shaking the pan often, sufficient boiling veal gravy
to almost cover them. Stew them gently from fifteen to sixteen
minutes, or longer should they not be perfectly tender. Add a
flavouring of mace, some salt, a quarter-pint of rich cream, a couple
of egg-yolks, and a little lemon-juice, observing, when the last are
added, the directions given for a blanquette of veal, page 229. Strips
of lemon-rind can be stewed in the gravy at pleasure. Two or three
dozens of mushroom-buttons, added twenty minutes before it is
served, will much improve this fricassée.
 SMALL ENTRÉES OF SWEETBREADS, CALF’S BRAINS AND
EARS, &c. &c.

For tables of which the service consists rather of a great variety of

light dishes (entrées) than of substantial English fare, the ears,
brains, sweetbreads, gristles or tendrons, and the tail of a calf, may
be dressed in many different ways to supply them; but they require a
really good style of cookery, and many adjuncts to render them
available for the purpose, as they do not possess much decided
natural flavour, and their insipidity would be apt to tire if it were not
relieved by the mode of preparing them. We shall give some few
especial receipts for them in the chapter on Foreign Cookery, should
sufficient space remain open for us to admit them; and insert here
only such slight general directions as may suffice for preparing some
of them in a simple form; as they are not in reality of first-rate
importance. All of them may be served with good curried, or highly-
flavoured tomato-sauce, after having been stewed in strong broth or
gravy. The brains and sweetbreads cut into small dice or scallops,
and mixed with béchamel, or with common white sauce, may be
used to fill small vol-au-vents, or patty cases. The ears are usually
filled in part with forcemeat, or a preparation of the brains, and
placed upright when dished; and the upper part is cut into narrow
fringe-like strips. For “Tendrons de Veau,” and “Breast of veal rolled
and stewed,” the reader is referred to Chapter XXXIV.
 CHAPTER XII.

Mutton.

No.
1. Leg.
2. Best End of Loin.
3. Chump End of Loin.
4. Neck, Best End.
5. Neck, Scrag End.
6. Shoulder.
7. Breast.
A Saddle is the Two Loins.
A Chine, the Two Necks.

Mutton is best suited for table in autumn, winter, and early spring. It is not
considered quite so good when grass-lamb is in full season, nor during the sultry
months of summer.
 TO CHOOSE MUTTON.

The best mutton is small-boned, plump, finely-grained, and short-

legged; the lean of a dark, rather than of a bright hue, and the fat
white and clear: when this is yellow, the meat is rank, and of bad
quality. Mutton is not considered by experienced judges to be in
perfection until it is nearly or quite five years old; but to avoid the
additional expense of feeding the animal so long, it is commonly
brought into the market at three years old. The leg and the loin are
the superior joints; and the preference would probably be given more
frequently to the latter, but for the superabundance of its fat, which
renders it a not very economical dish. The haunch consists of the leg
and the part of the loin adjoining it; the saddle, of the two loins
together, or of the undivided back of the sheep: these last are always
roasted, and are served usually at good tables, or for company-
dinners, instead of the smaller joints. The shoulder, dressed in the
ordinary way, is not very highly esteemed, but when boned, rolled,
and filled with forcemeat, it is of more presentable appearance, and
to many tastes, far better eating; though some persons prefer it in its
natural form, accompanied by stewed onions. It is occasionally
boiled or stewed, and covered with rich onion sauce. The flesh of
that part of the neck which is commonly called the “best end,” or the
back ribs, and which adjoins the loin, is the most succulent and
tender portion of the sheep, and makes an excellent small roast, and
is extremely good served as cutlets, after being divested of the
superabundant fat. It is likewise very frequently boiled; but so cooked
it makes but an unsightly and insipid dish, though an idea prevails in
this country that it is a very wholesome one. Cutlets (or chops, as the
butchers term them) are commonly taken from the loin, and are
generally charged at a higher rate than joints of mutton, in
consequence, probably, of the constant demand for them. They may
likewise be cut from the saddle, but will then be very large, and of no
better quality than when the two loins which form the saddle are
divided in the usual way, though a certain degree of fashion has of
late been accorded to them.[78] The scrag, or that part of it which
joins the head, is seldom used for any other purpose than making
broth, and should be taken off before the joint is dressed. Cutlets
from the thick end of the loin are commonly preferred to any others,
but they are frequently taken likewise from the best end of the neck
(sometimes called the back-ribs) and from the middle of the leg.
Mutton kidneys are dressed in various ways, and are excellent in
many. The trotters and the head of a sheep may be converted into
very good dishes, but they are scarcely worth the trouble which is
required to render them palatable. The loin and the leg are
occasionally cured and smoked like hams or bacon.
78. Many years since, these “saddle-back” cutlets were supplied to us by a
country butcher, and though of very fine South Down mutton, had no
particular importance attached to them, nor were they considered as
remarkably new.