Professional Documents
Culture Documents
Canadian Edition
Visit to download the full and correct content document:
https://ebooksecure.com/download/ebook-pdf-business-statistics-3rd-canadian-editio
n/
BUSINESS THIRD
CANADIAN
EDITION
STATISTICS
BUSINESS STATISTICS
SHARPE DE VEAUX
www.pearsoncanada.ca
ISBN 978-0-13-389912-2
SHARPE DE VEAUX
9 0 0 0 0
VELLEMAN WRIGHT
9 780133 899122
Meet the Authors
Norean Radke Sharpe (Ph.D. University of Virginia) has developed an international reputation
as an educator, administrator, and consultant on assessment and accreditation. She is currently a
professor at the McDonough School of Business at Georgetown University, where she is also Senior
Associate Dean and Director of Undergraduate Programs. Prior to joining Georgetown, Norean taught
Business Statistics and Operations Research courses to both undergraduate and MBA students for
14 years at Babson College. Before moving into business education, she taught Statistics for sev-
eral years at Bowdoin College and conducted research at Yale University. Norean is co-author of
the recent text, A Casebook for Business Statistics: Laboratories for Decision Making, and she has
authored more than 30 articles—primarily in the areas of Statistics education and women in science.
Norean currently serves as Associate Editor for the journal Cases in Business, Industry, and Govern-
ment Statistics. Her scholarship focuses on business forecasting, Statistics education, and student
learning. She is co-founder of the DOME Foundation, a nonprofit foundation that works to increase
diversity and outreach in mathematics and engineering, and she currently serves on two other non-
profit boards in the Washington, D.C., area. Norean has been active in increasing the participation of
women and underrepresented students in science and mathematics for several years and has two
children of her own.
Paul F. Velleman (Ph.D. Princeton University) has an international reputation for innovative
Statistics education. He designed the Data Desk® software package and is also the author and
designer of the award-winning ActivStats® multimedia software, for which he received the EDUCOM
Medal for innovative uses of computers in teaching Statistics and the ICTCM Award for Innovation
in Using Technology in College Mathematics. He is the founder and CEO of Data Description, Inc.
(www.datadesk.com), which supports both of these programs. He also developed the internet site Data
and Story Library (DASL; lib.stat.cmu.edu/DASL/), which provides data sets for teaching Statistics. Paul
co-authored (with David Hoaglin) the book ABCs of Exploratory Data Analysis. Paul teaches Statistics
at Cornell University in the Department of Statistical Sciences and in the School of Industrial and
Labor Relations, for which he has been awarded the MacIntyre Prize for Exemplary Teaching. His
research often focuses on statistical graphics and data analysis methods. Paul is a Fellow of the
American Statistical Association and of the American Association for the Advancement of Science.
Paul’s experience as a professor, entrepreneur, and business leader brings a unique perspective to
the book. Richard De Veaux and Paul Velleman have authored successful books in the introductory
college and AP High School market with David Bock, including Intro Stats, Fourth Edition (Pearson,
2014); Stats: Modeling the World, Fourth Edition (Pearson, 2015); and Stats: Data and Models, Third
Edition (Pearson, 2012).
David Wright combines an Engineering Ph.D. from Cambridge University, UK, with his current
position as Full Professor at the University of Ottawa’s Telfer School of Management to provide a
business perspective on new technology. Dr. Wright has taught in universities on three continents
and has experience in government and in industry. In government, he has developed statistical
models to evaluate the impact of industrial society on natural resource depletion. In industry, he
has worked with telecommunications equipment vendors and network operators on the impact
of new technology on business. His university teaching and research includes the economics of
solar energy, business statistics, and the environmental impact of the ICT industry. He is cited
in Who’s Who in the World, Who’s Who in Canadian Business, and Who’s Who in Science and
Engineering.
ix
Chapter 2 Data 8
2.1 What Are Data? 9 • 2.2 Variable Types 12 • 2.3 Where, How, and When 18
Ethics in Action 19
Mini Case Studies 21
Technology Help: Computer-Based Statistics Packages 22
xi
Appendixes
A. Answer Key A-1
B. Statistical Tables, Formulas, and Excel/XLStat B-1
C. Ethical Guidelines for Statistical Practice C-1
Index I-1
Communicating Results
shows a scatterplot of Calories and Sugar, coloured according to the shelf on which
the cereals were found, with a separate regression line fit for each shelf. Now we can
Are some months on the NYSE busier than others? Boxplots of the number of shares traded by month are a good
way to see such patterns. We’re interested not only in the centres, but also in the spreads. Are volumes equally
variable from month to month, or are they more spread out in some months?
see that the top shelf is unlike the bottom two shelves. We might want to report two
Plan/Do/Report
regressions, guided
one for the top shelf and oneexamples walk
for the bottom students
two shelves.1
through PLAN Setup Identify the variable, We want to compare the daily volume of shares traded from month to month
the steps of how to: report the time frame of the
data, and state the objective.
on the NYSE during 2006.
The daily volume is quantitative and measured in number of shares.
140
• Plan
120
the problem, checking assumptions and conditions. DO Mechanics Choose an appropri- We can partition the values by month and use side-by-side boxplots to com-
ate display for the data. pare the volume across months.
100
• Do80 the calculations. 3.0
Calories
2.7 *
• Report
40
findings within a business context. Bottom shelf
1.8
1.5
1.2
Middle shelf
0.9
20 Top shelf 0.6 *
1 2 3 4 5 6 7 8 9 10 11 12
Reports
0 are often presented
3 6
in the
9
form
12
of a memo to Month
–3 0
model realistic business Sugar
communications.
(g)
REPORT Conclusion Report what you’ve MEMO
Figure 19.5 Calories and Sugars coloured according to the shelf on which the cereal was learned about the data and any
Re: Research on the Trading Volume of the NYSE
found in a supermarket, with regression lines fit for each shelf individually. Do these data recommended action or analysis.
We have examined the daily sales volume on the NYSE (number of shares
appear homogeneous? That is, do all the cereals seem to be from the same population of traded) for each month of 2006. As the attached display shows, sales
cereals? Or are there kinds of cereals that we might want to consider separately? volume has lower median volume in March and August. The highest median
trading activity is found in November. The variability of trading volume
also shows a pattern. June and December have higher variability than the
rest, and March has noticeably less variability. There were several unusually
high-volume days that bear investigation and extremely low-volume days in
The price of a jewellery diamond depends on the 4 C’s: carat weight, cut, clarity, and
page 110
colour. Diamond colours are assigned letters from D, for colourless stones, through
K, for increasingly more yellow stones. A K-colour stone is only faintly yellow and still FOR EXAMPLE Comparing boxplots
considered jewellery quality. We’ve collected a sample of diamond prices from a website.
Here’s a scatterplot of Price vs. Carat Weight with points coloured according to Colour: QUESTION For the data in For Example: “Creating a histogram of the number of
For Example—short, illustrative business music downloads,” compare the a.m. downloads with the p.m. downloads by displaying
the two distributions side by side with boxplots.
Colour examples have been added ANSWER to most
There sections
are generally more downloadstoin the afternoon than in the morn-
ing. The median number of afternoon downloads is around 22 as compared with
10,000 D
E solidify understanding of newly learned statistical
14 for the morning hours. The p.m. downloads are also much more consistent. The
entire range of the p.m. hours, 15, is about the size of the IQR for a.m. hours. Both
8,000 F
G
methods and concepts.
6,000
Price
H
4,000 I
J
2,000
K
M05_SHAR9122_03_SE_C05.indd 110 29/08/16 5:53 PM
page 646
1
Another alternative is to fit a multiple regression model by adding variables (called dummy or
indicator variables) that distinguish the groups. This method will be discussed in Chapter 21.
Congratulations! You have now completed all seven chapters of Part I, “Exploring and Collecting Data.” A compre-
hensive case study available online in MyStatLab draws together many of the topics you have learned in Part I. Here
is a brief synopsis:
Case Studies covering
multiple chapters are provided
AIDS in Canada
at the end of Parts 1, 2, and
How do we tell the difference between a few isolated cases of a new disease and the start of an epidemic? This case
study puts you in the position of a manager at Health Canada, the Canadian Institute for Health Information, and/ 3 of the book, giving students
or the Public Health Agency of Canada when confronted by a new disease. It is based on real data that was available experience working through an
to such a person when AIDS was new and illustrates how to analyze unusual patterns and formats of data using the
information from Part I of this book. in-depth application using real 125
Mini Case Studies
case
MINI
M07_SHAR9122_03_SE_C07.indd 204
studies 20/09/16Exports
Canadian 5:33 PM
Statistics on Canadian exports are used for a variety of purposes, from pro-
jecting Canada’s foreign exchange earnings to planning capacity in Canadian
ports. The file ch05_MCSP_Canadian_Exports contains monthly
export data from Statistics Canada for three geographical areas.
Statistics Canada calculates exports on a “Customs” basis and also
on a “Balance of Payments” basis, and the file contains footnotes
Mini Case Studies at the end of each chapter ask students to describing the difference.6
investigate a question or make a business decision using real a) Draw time series graphs of this export data and identify any major
differences between the “Customs” and “Balance of Payments”
series.
data. Students define the objective, plan the process, complete b) Explain which basis of calculation, “Customs” or “Balance of
Payments,” would be appropriate for projecting Canada’s foreign
the analysis, and report their conclusion(s). Data are provided. exchange earnings.
c) Explain which basis of calculation, “Customs” or “Balance of Payments,” would be
appropriate for planning capacity in Canadian ports.
122 CHAPTER 5 Displaying and Describing Quantitative Data
d) Are there any exceptional periods during which exports in the three geographical
areas have differed from overall trends?
the summaries for you. After all, the computer doesn’t care what the numbers e) In order to forecast future exports, for which geographic region(s) would the mean
mean.
value of the data be appropriate? Give reasons for your answer.
• Watch out for multiple modes. If the distribution—as seen in a histogram, for page 125
example—has multiple modes, consider separating the data into groups. If
you can’t separate the data in a meaningful way, you shouldn’t summarize Solar Power in Ontario
the centre and spread of the variable. As a result of the Green Energy Act, Ontario has led the other Canadian provinces in
• Beware of outliers. If the data have outliers but are otherwise unimodal, con- solar power deployment. Many installers compete to mount solar modules on residen-
sider holding the outliers out of the further calculations and reporting them tial roofs in all the major cities, but an underserved market is small communities.
6
Based on Statistics Canada. CANSIM using CHASS, Table 228-0058: Merchandise imports and exports,
customs and balance of payments basis for all countries (2014) (dollars).
ETHICS IN ACTION
B
eth Ghazi owns Zenna’s Café, an independent coffee employees volunteered fewer than two hours per week,
shop located in a small city in Atlantic Canada. Since but Beth noticed that one part-time employee volunteered
opening Zenna’s in 2002, she has been steadily grow- more than 20 hours per week. She discovered that her What Can Go Wrong? 395
ing her business and now distributes her custom coffee employees collectively volunteered an average of 15 hours
blends to a number of regional restaurants and markets. She
operates a microroaster that offers specialty-grade Arabica
per month (with a median of eight hours). She planned to
report the average number and believed that most people
What Can Go
WHAT CAN M05_SHAR9122_03_SE_C05.indd
gO WRONg? 125 29/08/16 5:53 PM
coffees recognized as some of the best in the area. In addi-
tion to providing the highest-quality coffees, Beth wants her
would be impressed with Zenna’s level of commitment to
the local community.
Wrong? Sections
• Don’t base your null hypotheses on what you see in the data. You’re not allowed to
business to be socially responsible. To that end, she pays
fair prices to coffee farmers and donates profits to help
Ethical Issue The outlier in the data affects the average
in a direction that benefits Beth Ghazi and Zenna’s Café
look at the data first and then adjust your null hypothesis so that it will be
rejected. If your sample value turns out to be pn = 51.8% with a standard de-
near the end of each
chapter help students
charitable causes in Panama, Costa Rica, and Guatemala. viation of 1%, don’t form a null hypothesis as H0: p = 49.8%, knowing that
(related to Item C, ASA Ethical Guidelines; see Appendix C,
She also encourages her employees to get involved in the this will enable you to reject it. Your null hypothesis describes the “nothing
the American Statistical Association’s Ethical Guidelines for
local community. Recently, one of the well-known multina- interesting” or “nothing has changed” scenario and should not be based on
tional coffeehouse chains announced plans to locate shops
in her area. This chain is one of the few to offer Certified
Statistical Practice, also available online at www.amstat.org/
about/ethicalguidelines.cfm).
the data you collect. detect common
• Don’t base your alternative hypothesis on the data, either. You should always think
Free Trade coffee products and work toward social justice in
the global community.
Ethical solution Beth’s data are highly skewed. There is an
outlier value (for a part-time employee) that pulls the aver-
about the situation you’re investigating and base your alternative hypoth-
esis on that. Are you interested only in knowing whether something has
statistical errors and
Consequently, Beth thought it might be a good idea
for her to begin communicating Zenna’s message of
age number of volunteer hours up. Reporting the average
is misleading. In addition, there may be justification to
increased? Then write a one-tail (upper tail) alternative. Or would you be
equally interested in a change in either direction? Then you want a two- practise debunking
eliminate the value, since it belongs to a part-time employee tailed alternative. You should decide whether to do a one- or two-tailed test
misuses of statistics.
social responsibility to the public, but with an emphasis
on its commitment to the local community. Three months (and 10 of the 12 employees are full-time). It would be based on what results would be of interest to you, not on what you might
ago, she began collecting data on the number of volun- more ethical for Beth to (1) report the average but discuss see in the data.
teer hours donated by her employees per week. She has the outlier value; (2) report the average for only full-time • Don’t make your null hypothesis what you want to show to be true. Remember,
a total of 12 employees, of whom 10 are full-time. Most employees; or (3) report the median instead of the average. the null hypothesis is the status quo, the nothing-is-strange-here position
that a skeptic would take. You wonder whether the data cast doubt on that.
You can reject the null hypothesis, but you can never “accept” or “prove”
page 395
page 122 the null.
• Don’t forget to check the conditions. The reasoning of inference depends on
WHAT HAVE WE LEARNED? randomization. No amount of care in calculating a test result can save you
Ethics In
Learning Objectives
Action boxes in every chapter teach
➊ We’ve learned how to display and summarize quantitative data to help us see the story the
from a biased sample. The probabilities you compute depend on the Inde-
pendence Assumption. And your sample must be large enough to justify
Preface
The question that motivates a business student’s study of Statistics is “How can I make
better decisions?” As entrepreneurs and consultants, we know that in today’s data-rich envi-
ronment, knowledge of Statistics is essential to survive and thrive in the business world.
But, as educators, we’ve seen a disconnect between the way Business Statistics is tradition-
ally taught and the way it should be used in making business decisions. In Business Statistics,
we try to narrow the gap between theory and practice by presenting relevant statistical
methods that will empower business students to make effective, data-informed decisions.
Of course, students should come away from their Statistics course knowing how to
think statistically and how to apply Statistics methods with modern technology. But they
must also be able to communicate their analyses effectively to others. When asked about
Statistics education, a group of CEOs from Fortune 500 companies recently said that
although they were satisfied with the technical competence of students who had studied
Statistics, they found the students’ ability to communicate their findings to be woefully
inadequate.
Our “Plan, Do, Report” rubric provides a structure for solving business problems that
mimics the correct application of statistics to solving real business problems. Unlike many
other books, we emphasize the often neglected thinking (Plan) and communication (Report)
steps in problem solving in addition to the methodology (Do). This approach requires up-
to-date, real-world examples and data. So we constantly strive to illustrate our lessons with
current business issues and examples.
xix
• Section Exercises. Each chapter’s exercises begin with single-concept exercises that
target section topics. This makes it easier to check your understanding of each topic as
you learn it.
• Recent Data and New Examples. We teach with real data whenever possible. To keep
examples and exercises fresh, we’ve updated data throughout the book.
• Chapter Summaries. Our “What Have We Learned?” chapter summaries link back
to the Learning Objectives stated at the start of the chapter so that you can see clearly
when you have learned the key concepts in each chapter.
• Statistical Case Studies. Each chapter ends with a selection of Mini Case Studies, and
there are more extensive cases found on MyStatLab for Parts 1, 2, and 3 of the text.
Each part-ending case study covers material from multiple chapters.
• Streamlined “Technology Help” Sections with Additional Excel Coverage. Technol-
ogy Help sections were updated to ref lect changes in commercial software, and, when
possible, Excel screen shots are included to illustrate the menu options available.
• Emphasis on graphing and exploring data. Our consistent emphasis on the impor-
tance of displaying data is evident from the first chapters devoted to understanding
data to the sophisticated model-building chapters at the end of the book. Examples
often illustrate the value of examining data graphically, and the exercises reinforce
this concept. Graphics reveal structures, patterns, and occasional anomalies that could
otherwise go unnoticed. The sight of patterns displayed graphically often raises new
questions and informs both the path of a resulting statistical analysis and the ensuing
business decisions. The graphics that appear throughout the book also demonstrate
that the simple structures that underlie even the most sophisticated statistical infer-
ences are the same ones we look for in the simplest examples. That helps to tie the
concepts of the book together to tell a coherent story.
• Consistency. Having taught the importance of plotting data and checking assump-
tions and conditions, we are careful to model that behaviour throughout the book.
(Check the exercises in the chapters on multiple regression or time series and you’ll
find us still requiring and demonstrating the plots and checks that were introduced in
the early chapters.) This consistency helps reinforce these fundamental principles and
provides a familiar foundation for the more sophisticated topics.
Coverage
We were guided in our choice of topics by the GAISE (Guidelines for Assessment and
Instruction in Statistics Education) Report, which emerged from extensive studies explor-
ing how students best learn Statistics (www.amstat.org/education/gaise/). Those recommenda-
tions, now officially adopted and recommended by the American Statistical Association,
urge (among other detailed suggestions) that Statistics education should achieve the follow-
ing goals:
1. Emphasize statistical literacy and develop statistical thinking;
2. Use real data;
3. Stress conceptual understanding rather than mere knowledge of procedures;
4. Foster active learning;
5. Use technology for developing conceptual understanding and analyzing data; and
6. Use assessments to improve and evaluate student learning.
With respect to the order of topics, we followed the principle that a coherent introduc-
tory course should be designed so that concepts and methods fit together in a stepwise pro-
gression to provide a new understanding of how reasoning with data can uncover new and
important truths. For example, we teach inference concepts with proportions first and then
with means. Most students have had exposure to proportions through polls and advertising.
And by starting with proportions, we can teach inference with the Normal model and then
introduce inference for means with the Student’s t distribution. We introduce the concepts of
association, correlation, and regression early in Business Statistics. Our experience in the class-
room shows that introducing these fundamental ideas early makes Statistics useful and rele-
vant, even at the beginning of the course. Later in the semester, when we explore data through
inference, it feels natural and logical to build on the fundamental concepts learned earlier.
Syllabus Flexibility
Many instructors prefer to teach topics in a different sequence than the one presented in
the textbook. In order to assist you with your decision, Figure 1 is a diagram that illustrates
the dependency among chapters.
Ch 1–5 Ch 6 Ch 7
Data Correlation Regression
Ch 8–9
Probability and
Distributions
Ch 18–21
Ch 10–16 Ch 22–25
Regression
Inference Selected Topics
Inference
Ch 17
Nonparametrics
The subject of Business Statistics is sometimes taught in a single semester and other
times taught over the course of two semesters. Table 1 offers one suggestion for the way
in which chapters can be divided between two semesters.
Features
A textbook isn’t just words on a page—instead, it’s the cumulation of many features that
form a big picture. The features in Business Statistics are designed to provide a real-world
context for concepts, to help students to apply these concepts, to promote problem
solving, and to integrate technology—all in the name of helping students to more readily
identify the key themes the book is trying to teach.
Probability Multiple
Data Regression Distributions Inference Nonparametrics Regression Selected Topics
Motivating Vignettes. Each chapter opens with a motivating vignette, often taken
from the authors’ consulting experiences. These descriptions of companies—such as Bell
Canada, Sport Chek, Rogers, Intact Financial Corp., Ipsos Reid, PotashCorp of Saskatch-
ewan, Canada’s Wonderland, and Loblaw—enhance and illustrate the story of each chapter
and show how and why statistical thinking is so vital to modern business decision mak-
ing. We analyze data from or about the companies in the motivating vignettes throughout
the chapter.
For Examples. Nearly every section of every chapter includes a focused example that illus-
For Example trates and applies the concepts or methods of that section. The best way to understand and
remember a new theoretical concept or method is to see it applied in a real-world business
context right away. That’s what these examples do throughout the book.
Step-by-Step Guided Examples. The answer to a statistical question is almost never just a
PLAN number. Statistics is about understanding the world and making better decisions with data. To
that end, some examples in each chapter are presented as Guided Examples. A thorough solu-
tion is modelled in the right column while commentary appears in the left column. The over-
DO all analysis follows our innovative Plan, Do, Report template. That template begins each
analysis with a clear question about a business decision and an examination of the data avail-
able (Plan). It then moves to calculating the selected statistics (Do). Finally, it concludes with
REPORT a Report that specifically addresses the question. To emphasize that our goal is to address
the motivating question, we present the Report step as a business memo that summarizes
the results in the context of the example and states a recommendation if the data are able to
support one. To preserve the realism of the example, whenever it is appropriate we include
limitations of the analysis or models in the concluding memo, as would be required when
writing a report for management.
What Can Go What Can Go Wrong? Each chapter contains an innovative section called “What Can
Go Wrong?” which highlights the most common statistical errors and the misconceptions
Wrong? about Statistics. The most common mistakes for the new user of Statistics involve misusing
a method—not miscalculating a statistic. Most of the mistakes we discuss have been experi-
enced by the authors in a business context or in a classroom situation. One of our goals is
to arm students with the tools to detect statistical errors and to offer practice in debunking
misuses of Statistics, whether intentional or not. In this spirit, some of our exercises probe
how, and why, common errors tend to arise.
N otation A lert Notation Alert. Throughout this book, we emphasize the importance of clear communica-
tion. Proper notation is part of the vocabulary of Statistics, but it can be daunting. We all
know that in Algebra, n can stand for any variable, so it may be surprising to learn that in
Statistics, n is reserved for the sample size. Statisticians dedicate many letters and symbols
for specific meanings (e.g., the letters b, e, n, p, q, r, s, t, and z, along with many Greek let-
ters, all carry special connotations). Our “Notation Alerts” clarify which letters and sym-
bols statisticians use and the purpose of each letter and symbol.
Just Checking Just Checking. It is easy to start nodding in agreement without really understanding, so we
ask questions at points throughout the chapter. These questions are designed to conduct a
quick check of whether or not students have properly understood a section; most involve
very little calculation, and the answers are given in Appendix A. The questions can also be
used to motivate class discussion.
Optional Math Boxes. In many chapters we present the mathematical underpinnings of
Optional Math Box the statistical methods and concepts. We set proofs, derivations, and justifications apart
from the narrative in “Optional Math Boxes,” so the underlying mathematics is available
for those who want greater depth, but the text itself presents the logical development of the
topic at hand using a minimal amount of mathematics.
Ethics In Action Ethics in Action. Statistics involves more than simply plugging numbers into formulas;
most statistical analyses require a fair amount of judgment. When faced with these sorts of
important judgments, the best advice we can offer is to make an honest and ethical attempt
to address the appropriate business issue. The chapter-specific Ethics in Action boxes illus-
trate some of the judgments needed when conducting statistical analyses, identify possible
errors, link the issues to the American Statistical Association’s Ethical Guidelines, and then
propose ethically and statistically sound alternative approaches.
What Have Learning Objectives and What Have We Learned? Each chapter begins with a specific
list of learning objectives and ends by relating the objectives to the chapter summary (i.e.,
We Learned? the “What Have We Learned?” section). We review the concepts, define the terms intro-
duced in the chapter, and list the skills that form the core message of the chapter. The
“What Have We Learned?” sections make excellent study guides: the student who under-
stands the concepts in the summary, knows the terms, and practises the skills correctly is
better prepared to apply statistics to the world of business.
Technology Help Technology Help. At the end of each chapter, we summarize what students can find in the
most common software, often with annotated output. We then offer specific guidance for
Excel, Minitab, SPSS, and JMP, formatted in easy-to-read sections. This advice is intended
not to replace the documentation that accompanies the software, but rather to point the
way and provide startup assistance.
Mini Case Studies. Each chapter includes Mini Case Studies that ask students to conduct
an analysis based on a real business situation. Students define the objective, plan the pro-
case
MINI cess, complete the analysis, and report a conclusion. An ideal way for students to write up
studies
their work is the “Plan/Do/Report” format described above and used in each chapter. Data
for the Mini Case Studies are available on the MyStatLab site and are formatted for use
with various technologies.
Case Studies. Parts 1, 2, and 3 of the book have a Comprehensive Case Study on
MyStatLab. Students are given realistically large data sets (also on the MyStatLab site) and
challenged to respond to open-ended business questions using the data. Students have the
opportunity to bring together methods they have learned in the chapters included in that
part (and indeed, throughout the book) to address the issues raised. Students will be required
to use a computer to manipulate the large data sets that accompany these Case Studies.
Section Exercises. The Exercises for each chapter begin with a series of straightforward
Exer c i se s exercises targeted at the topics in each chapter section. This is the place to check under-
standing of specific topics. Because the exercises are labelled by section, turning back to the
right part of the chapter to clarify a concept or review a method is easy.
Chapter Exercises. These exercises are designed to be more realistic than the Section
Exercises and to lead to conclusions about practical management situations. The Chapter
Exercises may combine concepts and methods from different sections. We’ve worked hard
to make sure that they contain relevant, modern, and realistic business situations. When-
ever possible, the data are on the MyStatLab site (always in a variety of formats) so they can
be explored further. Often, we pair the exercises so that each odd-numbered exercise (with
answers that appear at the end of the book) is followed by an even-numbered exercise on
the same Statistics topic.