Applied Statistics in Business and

Economics, 7e ISE 7th Edition David

Doane
Visit to download the full and correct content document:
https://ebookmass.com/product/applied-statistics-in-business-and-economics-7e-ise-
7th-edition-david-doane/
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
Enhancements for Doane/Seward ASBE 7e
Many changes were motivated by advice from reviewers
and users of the textbook. Besides hundreds of small edits
and improved topic organization, these changes were com-
mon to most chapters:
∙ New overall design, colors, figures, and exercise layout
for a brighter and more efficient look.
∙ New end-of-chapter Software Supplements for R, includ-
ing two new appendixes (e.g., comparison of R with Excel)
Chapter 1—Overview of Statistics
Three new Analytics in Action (New AI Frontiers, Ethical
Issues in AI, Algorithms: Friend or Foe).
New Software Supplement (Introduction to R and RStudio).
Updated Related Readings.
Chapter 2—Data Collection
Updated examples and Super Bowl Mini Case.
New Analytics in Action (Too Much Randomness).
Revised discussion of scales, surveys, and sampling.
Chapter 3–Describing Data Visually
Leaner presentation of making charts.
Updated examples (e.g., U.S. foreign trade).
Updated seven exercises and new example (bank earnings).
Updated Related Readings.
New Software Supplement (Histograms and Scatter Plots
Using R).
Chapter 4–Descriptive Statistics
Streamlined discussion of main concepts.
Two new Analytics in Action (Future Job Titles, Measuring
Volatility).
Enhanced discussion of center and variability.
One new exercise (bank assets) and several updated exer-
cise data sets.
Updated Related Readings.
New Software Supplement (Descriptive Statistics Using R).
Chapter 5–Probability
Revised Venn diagram presentation.
New Analytics in Action (Climate Change).
Substantially revised, leaner treatment of contingency tables.
Two new exercises (clinical trials, 401K retirement
contributions).
New Software Supplement (Contingency Tables Using R).
∙ Updated test bank and updated/expanded Big Data Sets.
∙ Updated Related Readings and Web Sources for students
who want to “dive deeper.”
∙ Revised LearningStats demonstrations to illustrate
concepts beyond what is possible in a textbook (e.g.,
simulations).
∙ Improved illustrations, figures, and tables.
Chapter 6–Discrete Probability Distributions
One new Analytics in Action (Car Insurance Risk and
­Driverless Vehicles).
Several revised exercises.
New Software Supplement (Discrete Distributions Using R).
Chapter 7–Continuous Probability Distributions
Revised text, exercises, and illustrations.
New Software Supplement (Continuous Distributions
Using R).
Chapter 8–Sampling Distributions and Estimation
New introduction.
Revised treatment of CLT and deleted section 8.3.
New Analytics in Action (Margin of Error in the Era of
Big Data).
New discussion of bootstrap method.
New Software Supplement (Bootstrap Confidence Intervals
in Minitab Express).
New Software Supplement (Confidence Intervals Using R).
Chapter 9–One-Sample Hypothesis Tests
New Analytics in Action (Walmart, Big Data, and Retail
Analytics).
More on choosing α and β, significance and p-values.
Leaner discussion of power, OC curves, and variance tests.
New Software Supplement (One-Sample Hypothesis Tests
Using R).
Chapter 10—Two-Sample Hypothesis Tests
Leaner discussion of two-tailed tests and confidence intervals.
New Analytics in Action (A/B Testing: Old Technique, New
Application).
Updates in exercises and Related Readings.
New Software Supplement (Two-Sample Hypothesis Tests
Using R).
xix
Chapter 11—Analysis of Variance
New Analytics in Action (Experiments or Big Data?).
Leaner discussion of two-factor ANOVA.
Updated Related Readings.
New Software Supplement (ANOVA Using R).
Chapter 12—Simple Regression
New Analytics in Action (Predictive Maintenance and
Machine Learning).
Revised discussion of confidence and prediction intervals.
New MiniCase 12.4 (exports and imports).
Leaner discussion of ill-conditioned data and spurious
correlation.
New MiniCase 12.6 (assets and market capitalization).
Revised data set (U.S. price inflation) and updated Related
Readings.
New Software Supplement (Simple Regression Using R).
Chapter 13—Multiple Regression
Simplified introduction and revised treatment of confidence
and prediction intervals.
Delete Mini Case 13.4 (cockpit noise).
New Analytics in Action (People Analytics at Work).
New correlation matrix illustration (vehicle MPG), updated
data set (CPI changes) and a new data set (immunotherapy
drug prices).
Updated Related Readings.
New Software Supplement (Multiple Regression Using R).
Chapter 14—Time Series Analysis
Updated examples (U.S. labor Force, dollar exchange rates).
Updated examples of erratic (hurricanes, snowfall) and
consistent patterns (health spending, Amazon revenue).
Leaner trend-fitting presentation, new formula for com-
pound growth rate, and example of decomposition using R.
xx
Two new trend interpretation exercises.
New Analytics in Action (Trend? Or Bubble?).
Updated 16 exercise data sets (e.g., bird strikes, renewable
energy, PepsiCo, JetBlue, Coca-Cola revenue, revolving
debt, plane shipments, federal budget, Boston Marathon,
leisure industry, snowboarding, airspace delays).
Updated Related Readings.
New Software Supplement (Time Trends and Seasonality
Using R).
Chapter 15—Chi-Square Tests
Simplified examples of raw data conversion.
New Analytics in Action (Confusion Matrix for Machine
Learning).
Simplified treatment of GOF tests (Uniform, Normal, ECDF).
New exercise (age and social media preference).
Updated data sets (Derby, NL runs).
New Software Supplement (Chi-Square Tests Using R).
Chapter 16—Nonparametric Tests
One new exercise (movie reviews) and updated Related
Readings.
New Software Supplement (Nonparametric Tests Using R).
Chapter 17—Quality Management
New Analytics in Action (Big Data Tracks a Virus).
Updated discussion of acceptance sampling.
Updated discussion of software (e.g., R CRAN packages).
Updated Related Readings.
Chapter 18—Simulation
New table for random data in R.
Deleted bootstrap discussion.
Updated Related Readings.
 Final PDF to printer

Brief Contents
Chapter 1 Chapter 14
Overview of Statistics 2 Time-Series Analysis 578

Chapter 2 Chapter 15
Data Collection 24 Chi-Square Tests 624

Chapter 3 Chapter 16
Describing Data Visually 56 Nonparametric Tests 662

Chapter 4 Chapter 17
Descriptive Statistics 100 Quality Management 692

Chapter 5 Chapter 18
Probability 158 Simulation 18-1

Chapter 6 Appendixes
Discrete Probability Distributions 200 A Binomial Probabilities 734
B Poisson Probabilities 736
Chapter 7
C-1 Standard Normal Areas 739
Continuous Probability Distributions 238
C-2 Cumulative Standard Normal
Distribution 740
Chapter 8
D Student’s t Critical Values 742
Sampling Distributions and Estimation 278
E Chi-Square Critical Values 743
Chapter 9 F Critical Values of F.10 744
One-Sample Hypothesis Tests 322 G Solutions to Odd-Numbered Exercises 752
H Answers to Exam Review Questions 779
Chapter 10 I Writing and Presenting Reports 781
Two-Sample Hypothesis Tests 370 J Statistics in Excel and R 785
K Using R and RStudio 789
Chapter 11
Analysis of Variance 416 Index 797
Standard Normal Areas 815
Chapter 12
Cumulative Standard
Simple Regression 462
Normal Distribution 816
Student’s t Critical Values 818
Chapter 13
Multiple Regression 522
xxi

doa16287_fm_i-1.indd xxi 10/22/20 07:09 PM

Contents
Chapter 1 4.6 Covariance and Correlation 137
4.7 Grouped Data 141
Overview of Statistics 2 4.8 Skewness and Kurtosis 143
Chapter Summary 147
1.1 What is Statistics? 3
Chapter Exercises 149
1.2 Why Study Statistics? 5
1.3 Applying Statistics in Business 7
1.4 Statistical Challenges 10 Chapter 5
1.5 Critical Thinking 16
Chapter Summary 19 Probability 158
Chapter Exercises 19 5.1 Random Experiments 159
5.2 Probability 161
Chapter 2 5.3 Rules of Probability 165
5.4 Independent Events 170
Data Collection 24 5.5 Contingency Tables 174
2.1 Variables and Data 25 5.6 Tree Diagrams 181
2.2 Level of Measurement 29 5.7 Bayes’ Theorem 183
2.3 Sampling Concepts 33 5.8 Counting Rules 189
2.4 Sampling Methods 36 Chapter Summary 192
2.5 Data Sources 44 Chapter Exercises 193
2.6 Surveys 46
Chapter Summary 51 Chapter 6
Chapter Exercises 51
Discrete Probability Distributions 200
Chapter 3 6.1 Discrete Probability Distributions 201
6.2 Expected Value and Variance 204
Describing Data Visually 56 6.3 Uniform Distribution 208
3.1 Stem-and-Leaf Displays and Dot Plots 57 6.4 Binomial Distribution 210
3.2 Frequency Distributions and Histograms 61 6.5 Poisson Distribution 217
3.3 Effective Excel Charts 69 6.6 Hypergeometric Distribution 223
3.4 Line Charts 71 6.7 Geometric Distribution (Optional) 227
3.5 Column and Bar Charts 74 6.8 Transformations of Random Variables (Optional) 229
3.6 Pie Charts 78 Chapter Summary 232
3.7 Scatter Plots 80 Chapter Exercises 234
3.8 Tables 84
3.9 Deceptive Graphs 87 Chapter 7
Chapter Summary 90
Chapter Exercises 91 Continuous Probability
Distributions 238
Chapter 4 7.1 Continuous Probability Distributions 239
Descriptive Statistics 100 7.2 Uniform Continuous Distribution 241
7.3 Normal Distribution 243
4.1 Numerical Description 101 7.4 Standard Normal Distribution 246
4.2 Measures of Center 103 7.5 Normal Approximations 259
4.3 Measures of Variability 115 7.6 Exponential Distribution 263
4.4 Standardized Data 123 7.7 Triangular Distribution (Optional) 268
4.5 Percentiles, Quartiles, Chapter Summary 271
and Box Plots 128 Chapter Exercises 273

xxii
 Contents xxiii

Chapter 8 11.3 Multiple Comparisons 427

11.4 Tests for Homogeneity Of Variances 429
Sampling Distributions and 11.5 Two-Factor ANOVA Without Replication
Estimation 278 (Randomized Block Model) 433
11.6 Two-Factor ANOVA with Replication
8.1 Sampling and Estimation 279 (Full Factorial Model) 441
8.2 Central Limit Theorem 283 11.7 Higher-Order ANOVA Models (Optional) 449
8.3 Confidence Interval for a Mean (μ) With Chapter Summary 452
Known σ 287 Chapter Exercises 453
8.4 Confidence Interval for a Mean (μ) With
Unknown σ 291
8.5 Confidence Interval for a Proportion (π) 297 Chapter 12
8.6 Estimating from Finite Populations 304 Simple Regression 462
8.7 Sample Size Determination for a Mean 306
8.8 Sample Size Determination for a Proportion 308 12.1 Visual Displays and Correlation Analysis 463
8.9 Confidence Interval for a Population Variance, 12.2 Simple Regression 469
σ2 (Optional) 311 12.3 Regression Models 471
Chapter Summary 313 12.4 Ordinary Least Squares Formulas 475
Chapter Exercises 315 12.5 Tests for Significance 479
12.6 Analysis of Variance: Overall Fit 485
Chapter 9 12.7 Confidence and Prediction
Intervals for Y 489
One-Sample Hypothesis Tests 322 12.8 Residual Tests 492
9.1 Logic of Hypothesis Testing 323 12.9 Unusual Observations 498
9.2 Type I and Type II Errors 326 12.10 Other Regression Problems (Optional) 502
9.3 Decision Rules and Critical Values 330 12.11 Logistic Regression (Optional) 505
9.4 Testing a Mean: Known Population Variance 333 Chapter Summary 507
9.5 Testing a Mean: Unknown Population Variance 340 Chapter Exercises 509
9.6 Testing a Proportion 345
9.7 Power Curves and OC Curves (Optional) 354 Chapter 13
9.8 Tests for One Variance (Optional) 360
Chapter Summary 362 Multiple Regression 522
Chapter Exercises 364 13.1 Multiple Regression 523
13.2 Assessing Overall Fit 528
Chapter 10 13.3 Predictor Significance 532
13.4 Confidence Intervals for Y 536
Two-Sample Hypothesis Tests 370 13.5 Categorical Variables 538
10.1 Two-Sample Tests 371 13.6 Tests for Nonlinearity and Interaction 547
10.2 Comparing Two Means: Independent Samples 373 13.7 Multicollinearity 550
10.3 Confidence Interval for the Difference of 13.8 Regression Diagnostics 554
Two Means, μ1 − μ2 381 13.9 Other Regression Topics 560
10.4 Comparing Two Means: Paired Samples 383 13.10 Logistic Regression (Optional) 562
10.5 Comparing Two Proportions 389 Chapter Summary 564
10.6 Confidence Interval for the Difference of Two Chapter Exercises 566
Proportions, π1 − π2 397
10.7 Comparing Two Variances 398 Chapter 14
Chapter Summary 405
Chapter Exercises 406 Time-Series Analysis 578
14.1 Time-Series Components 579
Chapter 11 14.2 Trend Forecasting 583
14.3 Assessing Fit 596
Analysis of Variance 416 14.4 Moving Averages 598
11.1 Overview of ANOVA 417 14.5 Exponential Smoothing 600
11.2 One-Factor ANOVA (Completely Randomized 14.6 Seasonality 604
Model) 419 14.7 Index Numbers 609
 Final PDF to printer

xxiv Contents

14.8 Forecasting: Final Thoughts 611 17.7 Other Control Charts 711
Chapter Summary 612 17.8 Patterns in Control Charts 716
Chapter Exercises 614 17.9 Process Capability 718
17.10 Additional Quality Topics (Optional) 721
Chapter 15
Chapter Summary 725
Chapter Exercises 726
Chi-Square Tests 624
15.1 Chi-Square Test for Independence 625 Chapter 18
15.2 Chi-Square Tests for Goodness of Fit 636
15.3 Uniform Goodness-of-Fit Test 639 Simulation 18-1
15.4 Poisson Goodness-of-Fit Test 643
15.5 Normal Chi-Square Goodness-of-Fit Test 648 Appendixes
15.6 ECDF Tests (Optional) 651
Chapter Summary 652 A Binomial Probabilities 734
Chapter Exercises 653
B Poisson Probabilities 736
C-1 Standard Normal Areas 739
Chapter 16
C-2 Cumulative Standard Normal
Nonparametric Tests 662 Distribution 740
16.1 Why Use Nonparametric Tests? 663 D Student’s t Critical Values 742
16.2 One-Sample Runs Test 664
16.3 Wilcoxon Signed-Rank Test 667 E Chi-Square Critical Values 743
16.4 Wilcoxon Rank Sum Test 670 F Critical Values of F.10 744
16.5 Kruskal-Wallis Test for Independent Samples 673
16.6 Friedman Test for Related Samples 678 G Solutions to Odd-Numbered Exercises 752
16.7 Spearman Rank Correlation Test 681 H Answers to Exam Review Questions 779
Chapter Summary 684
Chapter Exercises 685 I Writing and Presenting Reports 781
J Statistics in Excel and R 785

Chapter 17 K Using R and RStudio 789

Quality Management 692

Index 797
17.1 Quality and Variation 693
17.2 Pioneers in Quality Management 695 Standard Normal Areas 815
17.3 Quality Improvement 697 Cumulative Standard
17.4 Control Charts: Overview 701
Normal Distribution 816
17.5 Control Charts for a Mean 702
17.6 Control Charts for a Range 710 Student’s t Critical Values 818

doa16287_fm_i-1.indd xxiv 10/22/20 07:09 PM

Applied Statistics
in Business and Economics
Seventh Edition
 1
CHAPTER

Overview of
Statistics
CHAPTER CONTENTS

1.1 What Is Statistics?

1.2 Why Study Statistics?
1.3 Applying Statistics in Business
1.4 Statistical Challenges
1.5 Critical Thinking

CHAPTER LEARNING OBJECTIVES

When you finish this chapter, you should be able to

LO 1-1 Define statistics and explain some of its uses.
LO 1-2 List reasons for a business student to study statistics.
LO 1-3 Explain the uses of statistics in business.
LO 1-4 State the common challenges facing business professionals using statistics.
LO 1-5 List and explain common statistical pitfalls.

2
W
hen managers are well informed about a com-
pany’s internal operations (e.g., sales, production,
inventory levels, time to market, warranty claims) and
competitive position (e.g., market share, customer satisfaction, repeat Juice Images/Getty Images

sales), they can take appropriate actions to improve their business. Managers
need reliable, timely information so they can analyze market trends and adjust
to changing market conditions. Better data also can help a company decide
which types of strategic information it should share with trusted business part-
ners to improve its supply chain. Statistics and statistical analysis permit data-
based decision making and reduce managers’ need to rely on guesswork.
Statistics is a key component of the field of business intelligence, which
encompasses all the technologies for collecting, storing, accessing, and ana-
lyzing data on the company’s operations in order to make better business
decisions. Statistics helps convert unstructured “raw” data (e.g., point-of-sale
data, customer spending patterns) into useful information through online ana-
lytical processing (OLAP) and data mining, terms that you may have encoun-
tered in your other business classes. Statistical analysis focuses attention on
key problems and guides discussion toward issues, not personalities or terri-
torial struggles. While powerful database software and query systems are the
key to managing a firm’s data warehouse, relatively small Excel spreadsheets
are often the focus of discussion among managers when it comes to “bottom
line” decisions. That is why Excel is featured prominently in this textbook.
In short, companies increasingly are using business analytics to support
decision making, to recognize anomalies that require tactical action, or to
gain strategic insight to align business processes with business objectives.
Answers to questions such as “How likely is this event?” or “What if this trend
continues?” will lead to appropriate actions. Businesses that combine mana-
gerial judgment with statistical analysis are more successful.

1.1 WHAT IS STATISTICS? LO 1-1

Statistics is the science of collecting, organizing, analyzing, interpreting, and presenting data. Define statistics and
Some experts prefer to call statistics data science, a trilogy of tasks involving data modeling, explain some of its uses.
analysis, and decision making. A statistic is a single measure, reported as a number, used

3
4 Applied Statistics in Business and Economics

to summarize a sample data set. Statistics may be thought of as a collection of methodolo-

gies to summarize, draw valid conclusions, and make predictions from empirical measure-
ments. Statistics helps us organize and present information and extract meaning from raw
data. Although it is often associated with the sciences and medicine, statistics is now used in
every academic field and every area of business.

Plural or Singular?
Statistics The science of collecting, organizing, analyzing, interpreting, and present-
ing data.
Statistic    A single measure, reported as a number, used to summarize a sample data
set.

Many different measures can be used to summarize data sets. You will learn throughout
this textbook that there can be different measures for different sets of data and different mea-
sures for different types of questions about the same data set. Consider, for example, a sample
data set that consists of heights of students in a university. There could be many different uses
for this data set. Perhaps the manufacturer of graduation gowns wants to know how long to
make the gowns; the best statistic for this would be the average height of the students. But an
architect designing a classroom building would want to know how high the doorways should
be and would base measurements on the maximum height of the students. Both the average
and the maximum are examples of a statistic.
You may not have a trained statistician in your organization, but any college graduate is
expected to know something about statistics, and anyone who creates graphs or interprets data
is “doing statistics” without an official title.
There are two primary kinds of statistics:
∙ Descriptive statistics refers to the collection, organization, presentation, and summary of
data (either using charts and graphs or using a numerical summary).
∙ Inferential statistics refers to generalizing from a sample to a population, estimating
unknown population parameters, drawing conclusions, and making decisions.
Figure 1.1 identifies the tasks and the text chapters for each.

Figure 1.1
Overview of Statistics

Statistics

Collecting and Making Inferences

Describing Data from Samples

Sampling Visual Numerical Probability Estimating Testing Regression Quality

and Surveys Displays Summaries Models Parameters Hypotheses and Trends Control
(Ch 2) (Ch 3) (Ch 4) (Chs 5–8) (Ch 8) (Chs 9–16) (Chs 12–14) (Ch 17)
Another random document with
no related content on Scribd:
 LIBRO SEXTO
DE LA DIANA DE
GEORGE DE
MONTEMAYOR

Despues que Arsileo se partio,

quedó Felismena con Amarilida la
pastora que con él estaua,
pidiendose vna a otra cuenta de
sus vidas, cosa muy natural de
las que en semejantes partes se
hallan. Y estando Felismena
contando a la pastora la causa de
su venida, llegó a la choça vn
pastor de muy gentil disposiçion y
arte: aunque la tristeza paresçia
que le traya encubierta gran parte
della. Quando Amarilida le vio,
con la mayor presteza que pudo
se leuantó para yrse, mas
Felismena la trauó de la saya,
sospechando lo que podia ser, y
le dixo: No sería justo (hermosa
pastora) que esse agrauio
reçebiesse de ti, quien tanto
desseo tiene de seruirte, como
yo. Mas como ella porfiasse de
yrse de alli, el pastor con muchas
lagrimas dezia: Amarilida, no
quiero que teniendo respecto a lo
que me haze suffrir, te duelas
deste desuenturado pastor, sino
que tengas cuenta con tu gran
valor y hermosura, y con que no
ay cosa en la uida que peor esté
a una pastora de tu qualidad, que
tratar mal a quien tanto la[1266]
quiere. Mira, Amarilida mia, estos
cansados ojos, que tantas
lagrimas han derramado, y uerás
la razon que los tuyos tienen de
no mostrarse ayrados contra este
sin uentura pastor. ¡Ay que me
huyes por no uer la razon que
tienes de aguardarme! Espera,
Amarilida, oyeme lo que digo, y
siquiera no me respondas. ¿Qué
te cuesta oyr a quien tanto le ha
costado uerte? Y boluiendose a
Felismena con muchas lagrimas
le pedia que no le dexasse yr: la
qual importunaua con muy
blandas palabras a la pastora,
que no tratasse tan mal a quien
mostraua quererla más que a sí: y
que le escuchasse pues en ello
auenturaua tan poco. Mas
Amarilida respondio: Hermosa
pastora, no me mandeys oyr a
quien dé más credito a sus
pensamientos que a mis palabras.
Cata que este que delante de ti
está, es uno de los desconfiados
pastores, que se sabe, y de los
que mayor trabajo dan a las
pastoras que quieren bien.
Filemon dixo contra Felismena:
Yo quiero (hermosa pastora) que
seas el juez entre mi y Amarilida,
y si yo tengo culpa del enojo que
comigo tiene, quiero perder la
vida. E si ella la tuuiera, no quiero
otra cosa, sino que en paga
desto, conozca lo que me deue.
De perder tú la vida (dixo
Amarilida) yo estoy bien segura,
porque ni a ti te quieres tanto mal,
que lo hagas, ni a mí tanto bien,
que por mi causa te pongas en
auentura de perder la vida. Mas
yo agora quiero, que esta
hermosa pastora juzgue, vista mi
razon y la tuya, quál es más digno
de culpa entre los dos. Sea assi
(diso Felismena) y sentemonos al
pie desta verde haya, junto al
prado florido que delante los ojos
tenemos, porque quiero ver la
razon, que cada vno tiene, de
quexarse del otro. Despues que
todos se vuieron assentado sobre
la uerde yerua, Filemon començo
a hablar desta manera: Hermosa
pastora, confiado estoy, que si
acaso has sido tocada de amores,
conoçeras la poca razon que
Amarilida tiene de quexarse de mí
y de sentir tan mal de la fe que le
tengo, que venga a ymaginar lo
que nadie de su pastor imaginó.
Has de saber, hermosa pastora,
que quando yo nasçi, y aun ante
mucho que nasçiesse, los hados
me destinaron para que amasse
esta hermosa pastora que delante
mis tristes y tus hermosos ojos
está, y a esta causa he
respondido con el effecto de tal
manera, que no creo que ay amor
como el mio, ni ingratitud como la
suya. Sucçedio, pues, que
seruiendola desde mi niñez, lo
mejor que yo he sabido, aurá
como çinco o seis meses, que mi
desuentura aportó por aqui a vn
pastor llamado Arsileo, el qual
buscaua vna pastora, que se
llama Belisa, que por çierto mal
suçesso, anda por estos bosques
desterrada. Y como fuesse tanta
su tristeza, sucçedio que esta
cruel pastora que aqui veys, o por
mançilla que tuuo dél, o por la
poca que tiene de mí, o por lo que
ella se sabe, jamas la he podido
apartar de su compañia. Y si
acaso le hablaua en ello paresçia
que me queria matar, porque
aquellos ojos que alli veys, no
causan menos espanto, quando
miran, estando ayrados, que
alegria, quando estan serenos.
Pues como yo estuuiesse tan
occupado, el coraçon de
grandissimo amor, el alma de vna
affeçion[1267] jamas oyda, el
entendimiento de los mayores
çelos, que nunca nadie tuuo,
quexauame a Arsileo con
sospiros, y a la tierra con amargo
llanto: mostrando la sin razon que
Amarilida me hazia. Ha le
causado tan grande
aborresçimiento auer yo
imaginado cosa contra su
honestidad que por vengarse de
mi, ha perseuerado en ello hasta
aora, y no tan solamente haze
esto, mas en viendome delante
sus ojos, se va huyendo como la
medrosa çierua de los
hambrientos lebreles. Ansi que
por lo que deues a ti misma, te
pido que juzgues, si es bastante
la causa que tiene de
aborresçerme y si mi culpa es tan
graue, que merezca por ella ser
aborresçido. Acabado Filemon de
dar cuenta de su mal, y de la sin
razon que su Amarilida le hazia,
la pastora Amarilida començo a
hablar desta manera: Hermosa
pastora, auerme Filemon, que ahi
está, querido bien (a lo menos
auerlo mostrado) sus seruicios an
sido tales, que me sería mal
contado dezir otra cosa; pero si
yo tambien he desechado, por
causa suya, el seruiçio de otros
muchos pastores, que por estos
valles repastan sus ganados, y
zagales a quien naturaleza no ha
dotado de menos graçia que a
otros, el mismo puede dezillo.
Porque las muchas uezes que yo
he sido requestada, y las que he
tenido la firmeza que a su fe
deuia, no creo que ha sido muy
lexos de su presençia, mas no
auia de ser esto parte para que él
me tuuiesse tan en poco que
ymaginasse de mí cosa contra lo
que a mí misma soy obligada;
porque si es ansi, y él lo sabe,
que a muchos que por mí se
perdian, yo he desechado por
amor dél, ¿cómo auia yo de
desechar a él por otro? ¿O
pensaua en él, o en mis amores?
Cien mil uezes me ha Filemon
açechado, no perdiendo pisada,
de las que el pastor Arsileo y yo
dauamos por este hermoso ualle,
mas él mismo diga si algun dia
oyó que Arsileo me dixesse cosa
que supiesse a amores, o si yo le
respondia alguna que lo
paresçiesse ¿Qué dia me vio
hablar Filemon con Arsileo, que
entendiesse de mis palabras otra
cosa, que consolalle de tan graue
mal como padesçia? Pues si esto
auia de ser causa que
sospechasse mal de su pastora,
¿quién mejor puede juzgarlo que
él mismo? Mira, hermosa Nimpha,
quan entregado estaua a
sospechas falsas y dudosas
ymaginaçiones, que jamas mis
palabras pudieron satisfazelle, ni
acabar con él que dexasse de
ausentarse deste ualle, pensando
él que con ausençia daria fin a
mis dias, y engañose, porque
antes me paresçe que lo dio al
contentamiento de los suyos. Y lo
bueno es que aun no se
contentaua Filemon de tener
çelos de mí, que tan libre estaua
como tú, hermosa pastora, aurás
entendido, más aun lo publicaua
en todas las fiestas, bayles,
luchas, que entre los pastores
desta sierra se hazian. Y esto ya
tú conosçes, si uenia en mayor
daño de mi honra que de su
contentamiento. En fin, él se
ausentó de mi presençia, y pues
tomó por mediçina de su mal cosa
que más se lo ha acresçentado,
no me culpe si me he sabido
mejor aprouechar del remedio de
lo que él ha sabido tomalle. Y
pues tú, hermosa pastora, as
uisto el contento que yo reçebi, en
que dixesses al desconsolado
Arsileo nueuas de su pastora, y
que yo misma fuy la que le
importuné que luego fuesse a
buscalla, claro está que no podia
auer entre los dos cosa de que
pudiessemos ser tan mal
juzgados, como este pastor
inconsideradamente nos ha
juzgado. Ansi que esta es la
causa de yo me auer resfriado del
amor que a Filemon tenía, y de no
me querer más poner a peligro de
sus falsas sospechas, pues me
ha traydo mi buena dicha a
tiempo, que sin forçarme a mí
misma, pudiesse muy bien
hazello. Despues que Amarilida
vuo mostrado la poca razon que
el pastor auia tenido de dar
credito a sus ymaginaçiones y la
libertad en que el tienpo le auia
puesto (cosa muy natural de
coraçones essentos), el pastor le
respondio desta manera: No
niego yo (Amarilida) que tu
bondad y discreçion no basta
para desculparte de qualquiera
sospecha. ¿Mas quieres tú por
uentura hazer nouedades en
amores, y ser inuentora de otros
nueuos effectos de los que hasta
agora auemos uisto? ¿Quándo
quiso bien vn amador, que
qualquiera occasion de çelos, por
pequeña que fuesse, no le
atormentasse el alma, quanto
más siendo tan grande como la
que tú con larga conuersaçion y
amistad de Arsileo me ha dado?
¿Piensas tú, Amarilida, que para
los çelos son menester
çertidumbres? Pues engañaste,
que las sospechas son las
prinçipales causas de tenellos.
Creer yo que querias bien a
Arsileo por via de amores, no era
mucho, pues el publicallo yo, tan
poco era de manera que tu honra
quedasse offendida: quanto más
que la fuerça de amor era tan
grande, que me hazia publicar el
mal de que me temia. Y puesto
caso que tu bondad me
assegurasse, quando a hurto de
mis sospechas la consideraua,
todavia tenía temor de lo que me
podia suçeder, si la conuersaçion
yua delante. Quanto a lo que
dizes que yo me ausenté, no lo
hize por darte pena, sino por uer
si en la mia podria auer algun
remedio, no uiendo delante mis
ojos a quien tan grande me la
daua, y tambien porque mis
importunidades no te la
causassen. Pues si en buscar
remedio para tan graue mal, fuy
contra lo que te deuia: ¿qué más
pena que la que tu ausençia me
hizo sentir? ¿O qué más muestra
de amor que no ser ella causa de
oluidarte? ¿Y qué mayor señal del
poco que comigo tenias, que
auelle tú perdido de todo punto
con mi ausençia? Si dizes que
jamas quisiste bien a Arsileo, aun
esso me da a mi mayor causa de
quexarme, pues por cosa en que
tan poco te yua, dexauas a quien
tanto te desseaua seruir. Ansi que
tanto mayor quexa tengo de ti,
quanto menos fue el amor que a
Arsileo has tenido. Estas son
(Amarilida) las razones, y otras
muchas que no digo, que en mi
fauor puedo traer: las quales no
quiero que me ualgan, pues en
caso de amores suelen ualer tan
poco. Solamente te pido que tu
çlemençia y la fe que sienpre te
he tenido, esten, pastora, de mi
parte, porque si ésta me falta, ni
en mis males podra auer fin, ni
medio en tu condiçion. Y con esto
el pastor dio fin a sus palabras, y
prinçipio a tantas lagrimas, que
bastaron juntamente con los
ruegos, y sentençia que en este
caso Felismena dio, para que el
duro coraçon de Amarilida se
ablandasse, y el enamorado
pastor boluiesse en graçia de su
pastora: de lo qual quedó tan
contento, como nunca jamas lo
estuuo, y aun Amarilida no poco
gozosa de auer mostrado quán
engañado estaua Filemon en las
sospechas que della tenía. Y
despues de auer passado alli
aquel dia con muy gran
contentamiento de los dos
confederados amadores, y con
mayor desassosiego de la
hermosa Felismena, ella otro dia
por la mañana se partio dellos,
despues de muy grandes
abraços, y prometimientos de
procurar siempre la una de saber
del buen sucçesso de la otra.
Pues Sireno muy libre del amor, y
Seluagia y Syluano muy más
enamorados que nunca, la
hermosa Diana muy descontenta
del triste sucçesso de su camino,
passaua la uida apasçentando su
ganado por la ribera del
caudaloso Ezla: adonde muchas
uezes, topandose unos a otros,
hablauan en lo que mayor
contento les daua. Y estando un
dia la discreta Seluagia con el su
Syluano junto a la fuente de los
alisos, llegó acaso la pastora
Diana, que uenia en busca de un
cordero que de la manada se le
auia huydo, el qual Syluano tenía
atado a un myrtho, porque
quando alli llegaron, le halló
beuiendo en la clara fuente, y por
la marca conosçio ser de la
hermosa Diana. Pues siendo,
como digo, llegada y resçebida de
los dos nueuos amantes, con
gran cortesia se assento entre la
uerde yerua, arrimada a uno de
los alisos que la fuente rodeauan,
y despues de auer hablado en
muchas cosas, le dixo Syluano:
¿Cómo (hermosa Diana) no nos
preguntas por Sireno? Diana
entonces le respondio: Como no
querria tratar de cosas passadas,
por lo mucho que me fatigan las
presentes: tienpo fue que
preguntar yo por él le diera más
contento, y aun a mí el hablalle,
de lo que a ninguno de los dos
aora nos dara, mas el tienpo cura
infinitas cosas que a la persona le
paresçen sin remedio. Y si esto
assi no entendiesse, ya no auria
Diana en el mundo, segun los
desgustos y pesadumbres que
cada dia se me offreçen. No
querra Dios tanto mal al mundo
(respondio Seluagia), que le quite
tan grande hermosura como la
tuya. Essa no le faltará en quanto
tú biuieres (dixo Diana) y adonde
está tu graçia y gentileza muy
poco se perderia en mí. Sino
miralo por el tu Syluano, que
jamas pensé yo que él me
oluidara por otra pastora alguna, y
en fin me ha dado de mano por
amor de ti. Esto dezia Diana, con
una risa muy graçiosa, aunque no
se reya destas cosas tanto, ni tan
de gana, como ellos pensauan.
Que puesto caso que ella uuiesse
querido a Sireno más que a su
uida, y a Syluano le uuiesse
aborresçido, más le pesaua del
oluido de Syluano, por ser
causado de otra, de cuya vista
estaua cada dia gozando con
gran contentamiento de sus
amores, que del oluido de Sireno,
a quien no mouia ningun
pensamiento nueuo. Quando
Syluano oyó lo que Diana auia
dicho, le respondio: Oluidarte yo,
Diana, seria escusado, porque no
es tu hermosura y ualor de los
que oluidarse pueden. Verdad es
que yo soy de la mi Seluagia:
porque de más de auer en ella
muchas partes, que hazello me
obligan, no tuuo en menos su
suerte, por ser amada de aquél a
quien tú en tan poco tuuiste.
Dexemos esso (dixo Diana) que
tú estás muy bien empleado, y yo
no lo miré bien, en no quererte
como tu amor me lo meresçia. Si
algun contento en algun tienpo
desseaste darme, ruegote todo
quanto puedo que tú y la hermosa
Seluagia canteys alguna cançion
por entretener la fiesta: que me
paresçe que comiença de manera
que será forçado passalla debaxo
de estos alisos, gustando del
ruydo de la clara fuente, el qual
no ayudará poco a la suavidad de
vuestro canto. No se hizieron de
rogar los nueuos amadores,
aunque la hermosa Seluagia no
gustó mucho de la platica que
Diana con Syluano auia tenido.
Mas porque en la cançion pensó
satisfazer al son de la çampoña
que Diana tañía, començaron los
dos a cantar desta manera:

Zagal alegre te ueo,

y tu fe firme y segura.
—Cortome amor la uentura
a medida del desseo.
¿Qué desseaste alcançar,
que tal contento te diesse?
—Querer a quien me
quisiesse,
 que no hay más que dessear.
Essa gloria en que te ueo,
tienes la por muy segura.
—No me la ha dado uentura
para burlar al desseo.
¿En quanto estuuiese
firme[1268],
moririas sospirando?
—De oyllo dezir burlando
estoy ya para morirme.
¿Mudarias (aunque feo)
viendo mayor hermosura?
—No porque sería locura
pedirme más el desseo.
¿Tienesme tan grande
amor,
como en tus palabras siento?
—Esso a tu meresçimiento
lo preguntarás mejor.
Algunas uezes lo creo,
y otras no estoy muy segura.
—Solo en eso la uentura
haze offensa a mi desseo.
Finge que de otra zagala
te enamoras más hermosa.
—No me mandes hazer cosa,
que aun para fingida es mala.
Muy más firmeza te ueo,
pastor, que a mi hermosura.
—Y a mí muy mayor uentura
que jamas cupo en desseo.

A este tiempo baxaua Sireno del

aldea, á la fuente de los alisos,
con grandissimo desseo de topar
a Seluagia, o a Syluano. Porque
ninguna cosa por entonçes le
daua más contento que la
conuersaçion de los dos nueuos
enamorados. Y paseando por la
memoria los amores de Diana, no
dexaua de causalle soledad el
tiempo que la auia querido, no
porque entonçes le diesse pena
su amor, mas porque en todo
tienpo la memoria de un buen
estado causa soledad al que le ha
perdido. Y antes que llegasse a la
fuente, en medio del uerde prado,
que de myrthos y laureles
rodeado estaua, halló las ouejas
de Diana, que solas por entre los
arboles andauan pasçiendo, so el
amparo de los brauos mastines. Y
como el pastor se parasse a
mirallas, ymaginando el tienpo en
que le auian dado más en que
entender que las suyas proprias;
los mastines con gran furia se
uinieron a él, mas como llegassen
y dellos fuesse conosçido,
meneando las colas y baxando
los pescueços que de agudas
puntas de azero estauan
rodeados, se le echaron a los
pies, y otros se empinauan con el
mayor regozijo del mundo. Pues
las ouejas no menos sentimiento
hizieron, porque la borrega mayor,
con su rustico çençerro, se uino al
pastor, y todas las otras guiadas
por ella, o por el conosçimiento de
Sireno, le çercaron alrededor,
cosa que él no pudo uer sin
lagrimas, acordandosele que en
compañía de la hermosa pastora
Diana auia repastado aquel
rebaño. Y uiendo que en los
animales sobraua el
conosçimiento que en su señora
auia faltado, cosa fue ésta, que si
la fuerça del agua que la sabia
Feliçia le auia dado, no le uuiera
hecho oluidar los amores, quiça
no uuiera cosa en el mundo que
le estoruara boluer a ellos. Mas
uiendose çercado de las ouejas
de Diana, y de los pensamientos
que la memoria della ante los ojos
le ponia, començo a cantar esta
cançion al son de su loçano rabel.

Passados contentamientos
¿qué quereys?
dexadme, no me canseys.

Memoria, ¿quereys oyrme?

los dias, las noches buenas,
paguelos con las setenas,
no teneys más que pedirme,
todo se acabó en partirme,
como ueys,
dexadme, no me canseys.

Campo uerde, ualle

vmbroso,
donde algun tiempo gozé,
ved lo que despues passé,
y dexadme en mi reposo:
si estoy con razon medroso,
ya lo ueys,
 dexadme, no me canseys.

Vi mudado un coraçon,
cansado de assegurarme,
fue forçado aprouecharme,
del tiempo, y de la occasion;
memoria do no ay passion,
¿qué quereys?
dexadme, no me canseys.

Corderos y ouejas mias,

pues algun tiempo lo fuistes,
las horas lentas o tristes
passaronse con los dias,
no hagays las alegrias
que soleys,
pues ya no me engañareys.

Si uenis por me turbar,

si uenis por consolar,
ya no hay mal que consolar:
si uenis por me matar,
bien podeys,
matadme y acabareys.

Despues que Sireno vuo cantado,

en la boz fue conosçido de la
hermosa pastora Diana y de los
dos enamorados, Seluagia y
Syluano. Ellos le dieron bozes,
diziendo que si pensaua passar la
fiesta en el campo, que alli estaua
la sabrosa fuente de los alisos, y
la hermosa pastora Diana, que no
seria mal entretenimiento para
passalla. Sireno le respondio que
por fuerça auia de esperar todo el
dia en el campo, hasta que fuesse
hora de boluer con el ganado a su
aldea, y viniendose adonde el
pastor y pastoras estauan, se
sentaron en torno de la clara
fuente, como otras uezes solian.
Diana, cuya uida era tan triste
qual puede ymaginar quien uiesse
una pastora la más hermosa y
discreta que entonces se sabia,
tan fuera de su gusto casada,
siempre andaua buscando
entretenimientos para passar la
uida hurtando el cuerpo a sus
imaginaçiones. Pues estando los
dos pastores hablando en
algunas cosas tocantes al pasto
de los ganados y al
aprouechamiento dellos, Diana
les rompio el hilo de su platica,
diziendo contra Syluano: Buena
cosa es, pastor, que estando
delante la hermosa Seluagia
trates de otra cosa, sino de
encaresçer su hermosura y el
gran amor que te tiene: dexa el
campo, y los corderos, los malos,
o buenos sucçessos del tiempo y
fortuna, y goza, pastor, de la
buena que has tenido, en ser
amado de tan hermosa pastora,
que adonde el contentamiento del
spirito es razon que sea tan
grande, poco al caso hazen los
bienes de fortuna. Siluano
entonces le respondio: Lo mucho
que yo, Diana, te deuo, nadie lo
sabría encaresçer, como ello es,
sino quien huuiese entendido la
razon que tengo de conoçer esta
deuda, pues no tan solo me
enseñaste a querer bien, mas aun
aora me guyas y muestras vsar
del contentamiento que mis
amores me dan. Infinita es la
razon que tienes de mandarme
que no trate de otra cosa, estando
mi señora delante, sino del
contento que su vista me causa, y
assi prometo de hacello, en
quanto el alma no se despidiere
destos cansados miembros. Mas
de una cosa estoy espantado, y
es de ver como el tu Sireno
buelue a otra parte los ojos,
quando hablas; paresçe, que no
le agradan tus palabras, ni se
satisfaçe de lo que respondes. No
le pongas culpa (dixo Diana) que
hombres descuydados y
enemigos de lo que a si mismos
deuen, esso y más harán.
¿Enemigo de lo que a mí mismo
deuo? (respondia Sireno). Si yo
jamas lo fuy, la muerte me dé la
pena de mi yerro. Buena manera
es essa de desculparte.
¡Desculparme yo, Sireno (dixo
Diana) si la primera culpa contra ti
no tengo por cometer, jamas me
vea con más contento, que el que
agora tengo! Bueno es que me
pongas tú culpa por auerme
casado, teniendo padres. Mas
bueno es (dixo Sireno) que
casasses teniendo amor. ¿Y qué
parte (dixo Diana) era el amor,
adonde estaua la obediencia que
a los padres se deuia? ¿Mas qué
parte (respondio Sireno) eran los
padres, la obediençia, los
tiempos, ni los malos ó fauorables
sucçessos de la fortuna, para
sobrepujar vn amor tan
verdadero, como antes de mi
partida me mostraste? Ah Diana,
Diana, que nunca yo pense que
vuiera cosa en la uida que vna fe
tan grande pudiera quebrar:
quanto más, Diana, que bien te
pudieras casar, y no olvidar a
quien tanto te queria. Mas
mirandolo desapassionadamente,
muy mejor fue para mí ya que te
casauas, el oluidarme. ¿Por qué
razon (dixo Diana?) Porque no ay
(respondio Sireno) peor estado
que es querer vn pastor á una
pastora casada: ni cosa que más
haga perder el seso, al que
uerdadero amor le tiene. Y la
razon dello es, que como todos
sabemos, la principal passion,
que a un amador atormenta,
despues del desseo de su dama
son los çelos. Pues qué te
paresçe, que será para un
desdichado que quiere bien,
saber que su pastora está en
braços de su uelado, y él llorando
en la calle su desuentura: Y no