Business Statistics Communicating With Numbers 4E 4Th Edition Sanjiv Jaggia Full Chapter

Business Statistics: Communicating
with Numbers, 4e 4th Edition Sanjiv

Jaggia
Visit to download the full and correct content document:
https://ebookmass.com/product/business-statistics-communicating-with-numbers-4e-
4th-edition-sanjiv-jaggia/
ISTUDY
BUSINESS STATISTICS
ISTUDY
The McGraw Hill Series in Operations and Decision Sciences
SUPPLY CHAIN MANAGEMENT BUSINESS SYSTEMS DYNAMICS Doane and Seward
Essential Statistics in Business and
Bowersox, Closs, Cooper, and Bowersox Sterman
Economics
Supply Chain Logistics Management Business Dynamics: Systems Thinking
Third Edition
Fifth Edition and Modeling for a Complex World
Lind, Marchal, and Wathen
Johnson OPERATIONS MANAGEMENT
Basic Statistics for Business and
Purchasing and Supply Management
Cachon and Terwiesch Economics
Sixteenth Edition
Operations Management Tenth Edition
Simchi-Levi, Kaminsky, and Simchi-Levi Second Edition
Lind, Marchal, and Wathen
Designing and Managing the Supply
Cachon and Terwiesch Statistical Techniques in Business and
Chain: Concepts, Strategies, Case
Matching Supply with Demand: An Economics
Studies
Introduction to Operations Management Eighteenth Edition
Fourth Edition
Fourth Edition
Jaggia and Kelly
Stock and Manrodt
Jacobs and Chase Business Statistics: Communicating
Fundamentals of Supply Chain
Operations and Supply Chain with Numbers
Management
Management Fourth Edition
PROJECT MANAGEMENT Sixteenth Edition
Jaggia and Kelly
Larson and Gray Jacobs and Chase Essentials of Business Statistics:
Project Management: The Managerial Operations and Supply Chain Communicating with Numbers
Process Management: The Core Second Edition
Eighth Edition Fifth Edition
BUSINESS ANALYTICS
SERVICE OPERATIONS Schroeder and Goldstein
Jaggia, Kelly, Lertwachara, and Chen
MANAGEMENT Operations Management in the Supply
Business Analytics: Communicating
Chain: Decisions and Cases
Bordoloi, Fitzsimmons, and Fitzsimmons with Numbers
Eighth Edition
Service Management: Operations,
BUSINESS MATH
Strategy, Information Technology Stevenson
Ninth Edition Operations Management Slater and Wittry
Fourteenth Edition Practical Business Math Procedures
MANAGEMENT SCIENCE
Thirteenth Edition
Swink, Melnyk, and Hartley
Hillier and Hillier
Managing Operations Across the Supply Slater and Wittry
Introduction to Management Science: A
Chain Math for Business and Finance: An
Modeling and Case Studies Approach
Fourth Edition Algebraic Approach
with Spreadsheets
Second Edition
Sixth Edition BUSINESS STATISTICS
BUSINESS RESEARCH METHODS Bowerman, Drougas, Duckworth, Froelich,
Hummel, Moninger, and Schur
Schindler
Business Statistics and Analytics in
Business Research Methods
Practice
Fourteenth Edition
Ninth Edition
BUSINESS FORECASTING
Doane and Seward
Keating and Wilson Applied Statistics in Business and
Forecasting and Predictive Analytics Economics
Seventh Edition Seventh Edition
ISTUDY
Fourth Edition
BUSINESS STATISTICS
Communicating with Numbers
Sanjiv Jaggia Alison Kelly

California Polytechnic Suffolk University
State University
ISTUDY
Final PDF to printer
BUSINESS 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 978-1-260-59756-1
MHID 1-260-59756-3
Cover Image: McGraw Hill
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
ISTUDY jag97563_fm_ise iv 11/07/20 12:06 PM

Dedicated to Chandrika, Minori, John, Megan, and Matthew
ISTUDY
A B O U T T H E AU T H O R S
Sanjiv Jaggia
Sanjiv Jaggia is a professor of economics and finance at
California Polytechnic State University in San Luis Obispo.
Dr. Jaggia holds a Ph.D. from Indiana University and is a
Chartered Financial Analyst (CFA®). He enjoys research
in statistics and data analytics applied to a wide range of
business disciplines. Dr. Jaggia has published numerous
papers in leading academic journals and has co-authored
three successful textbooks, two in business statistics and
Courtesy Sanjiv Jaggia
one in business analytics. His ability to communicate in the
classroom has been acknowledged by several teaching
awards. Dr. Jaggia resides in San Luis Obispo with his wife and daughter. In his spare
time, he enjoys cooking, hiking, and listening to a wide range of music.
Alison Kelly
Alison Kelly is a professor of economics at Suffolk University in Boston. Dr. Kelly holds a Ph.D.
from Boston College and is a Chartered Financial Analyst (CFA®). Dr. Kelly has published in
a wide variety of academic journals and has co-authored three successful textbooks, two
in business statistics and one in business analytics. Her
courses in applied statistics and econometrics are popular
with students as well as working professionals. She has
also served as a consultant for a number of companies;
her most recent work focused on how large financial
institutions satisfy requirements mandated by the Dodd-
Frank Act. Dr. Kelly resides in Hamilton, Massachusetts,
with her husband, daughter, and son. In her spare time,
Courtesy Alison Kelly
she enjoys exercising and gardening.
vi
ISTUDY
Business Statistics: Communicating
with Numbers
Reviewer Quotes
“[Jaggia and Kelly’s text is] an introductory statistics textbook which is

rigorous in statistics with modern technology embedded.”
-Qiang Zhen, University of North Florida
“This introductory statistics book is relevant and approachable. The

book and its materials support teaching in various modalities. It offers
an applied orientation with a reasonable and appropriate theoretical
foundation.”
-Kathryn Ernstberger, Indiana University Southeast
“The authors . . . do an excellent job of introducing the concepts.

Illustrations, modern and relevant examples and applications,
and exercises. . . This is a well-rounded and excellent textbook in
introductory statistics and data analysis.”
-Mohammad A. Kazemi, University of North Carolina, Charlotte
“This book and its accompanying online resources is ideal for an

introductory in statistics with an emphasis in business. There is a
thoughtful balance between concepts and applications.”
-Ted Galanthay, Ithaca College
“Excellent coverage. . . It is a great book.”

-Ricardo S. Tovar-Silos, Lamar University
WALKTHROUGH
ISTUDY
A Unique Emphasis on
Communicating with Numbers
Makes Business Statistics Relevant
to Students
We wrote Business Statistics: Communicating with Numbers because we saw a need
for a contemporary, core statistics text that sparked student interest and bridged the gap
between how statistics is taught and how practitioners think about and apply statistical
methods. Throughout the text, the emphasis is on communicating with numbers rather
than on number crunching. In every chapter, students are exposed to statistical informa-
tion conveyed in written form. By incorporating the perspective of practitioners, it has
been our goal to make the subject matter more relevant and the presentation of material
more straightforward for students. Although the text is application-oriented and practical,
it is also mathematically sound and uses notation that is generally accepted for the topic
being covered.
From our years of experience in the classroom, we have found that an effective way
to make statistics interesting is to use timely applications. For these reasons, examples
in Business Statistics come from all walks of life, including business, economics, sports,
health, housing, the environment, polling, and psychology. By carefully matching exam-
ples with statistical methods, students learn to appreciate the relevance of statistics in our
world today, and perhaps, end up learning statistics without realizing they are doing so.
Continuing Key Features

The fourth edition of Business Statistics reinforces and expands six core features that
were well-received in earlier editions.
Integrated Introductory Cases. Each chapter begins with an interesting and relevant
introductory case. The case is threaded throughout the chapter, and once the relevant sta-
tistical tools have been covered, a synopsis—a short summary of findings—is provided.
The introductory case often serves as the basis of several examples in other chapters.
Writing with Data. Interpreting results and conveying information effectively is critical
to effective decision making in a business environment. Students are taught how to take
the data, apply it, and convey the information in a meaningful way.
Unique Coverage of Regression Analysis. Relevant coverage of regression without

repetition is an important hallmark of this text.
Written as Taught. Topics are presented the way they are taught in class, beginning
with the intuition and explanation and concluding with the application.
Integration of Microsoft Excel® and R. Students are taught to develop an understand-

ing of the concepts and how to derive the calculation; then Excel and R are used as a tool
to perform the cumbersome calculations.
Connect. Connect is an online system that gives students the tools they need to be suc-
cessful in the course. Through guided examples and LearnSmart adaptive study tools,
students receive guidance and practice to help them master the topics.
viii B U S I N E S S S TAT I S T I C S WALKTHROUGH
ISTUDY
Features New to the Fourth Edition
In the fourth edition of Business Statistics, we have made substantial revisions that
address the current needs of the market. These revisions are based on the feedback of
countless reviewers and users of our earlier editions.
The emphasis in this edition has been to strengthen the connection between business
statistics and data analytics. More than ever, colleges and universities across the United
States and abroad are incorporating business analytics into their curricula, and businesses
are scrambling to find qualified professionals who can translate statistical analysis into
decisions that improve performance. We believe that the fourth edition will not only
introduce students to data analytics, but will also excite them to further explore the field.
There are four major innovations in this edition.
Descriptive: More emphasis on data preparation and visualization

∙ New sections devoted to data preparation in Chapter 1
∙ New sections devoted to data visualization methods in Chapter 2
∙ Discussion of subsetted means in Chapter 3
∙ Discussion of pivot tables used to analyze empirical probabilities in Chapter 4
Predictive: Significant rewrite of regression and forecasting

∙ Streamlined discussion of goodness-of-fit measures in Chapter 14
∙ Revised section on model assumptions and common violations in Chapter 15
∙ Improved visualizations to explore nonlinear models in Chapters 16, 17, and 18
∙ New subsection on accuracy rates for binary choice models in Chapter 17
∙ Separate sections devoted to linear and nonlinear forecasting models in Chapter 18
Technology: More reliance on statistical software and Connect

∙ Greater use of Excel and R in solving problems
∙ Expanded R instructions in all regression and forecasting chapters
∙ Excel and/or R instructions included in most exercises in Connect
∙ Improved Connect product to facilitate teaching in an online environment
Storytelling: More relevant discussion

∙ Numerous new examples, exercises, and case studies
∙ Updated data to make the applications more current
∙ Big data used in the writing sections for Chapters 1, 2, 3, 15, 16, and 17
∙ Big data used for suggested case studies in several chapters
WALKTHROUGH
ISTUDY
Students Learn Through Real-World
Cases and Business Examples . . . Rev. Confirming Pages
Integrated Introductory Cases The numerator of the Sharpe ratio measures the extra reward that investors receive for the
added risk taken—this difference is often called excess return. The higher the Sharpe
ratio, the better the investment compensates its investors for risk.
Each chapter opens with a real-life case study that forms the basis for several examples
within the chapter. The questions included in the
EXAMPLE examples
3.19 create a roadmap for master-
ing the most important learning outcomes within the chapter. A onsynopsis
Growth and Value given that the return of each chap-
Use the information in Table 3.13 to calculate and interpret the Sharpe ratios for
a 1-year T-bill is 2%.
ter’s introductory case is presented when the last

SOLUTIONof these
Since the returnexamples
on a 1-year T-bill ishas
2%, R¯ =been discussed.
2. Plugging in the values
of the relevant means and standard deviations into the Sharpe ratio yields
f
Instructors of distance learners may find these introductory_____ cases

x¯ − R¯
particularly
15.755 − 2
_________
useful. I f
Sharpe ratio for Growth : = = 0.58.
sI 23.799
x¯ I − R¯ f _________
_____ 12.005 − 2
Sharpe ratio for Value: = = 0.56.
sI 17.979
Rev. Confirming Pages
We had earlier shown that Growth had a higher return, which is good, along with
a higher variance, which is bad. We can use the Sharpe ratio to make a valid com-
parison between the mutual funds. Growth provides a higher Sharpe ratio than Value
(0.58 > 0.56); therefore, the Growth offered more reward per unit of risk compared
to Value.
SYNOPSIS OF INTRODUCTORY CASE

Growth and value are two fundamental styles in stock and mutual
fund investing. proponents of growth investing believe that compa-
nies that are growing faster than their peers are trendsetters and
will be able to maintain their superior growth. By investing in the
stocks of these companies, they expect their investment to grow
at a rate faster than the overall stock market. By comparison, value
investors focus on the stocks of companies that are trading at a
discount relative to the overall market or a specific sector. Investors
of value stocks believe that these stocks are undervalued and that
their price will increase once their true value is recognized by other
investors. The debate between growth and value investing is age- Gladkikh/Getty Images
old, and which style dominates depends on the sample period used
for the analysis.
An analysis of annual return data for Fidelity’s Growth Index mutual fund (Growth) and Fidelity’s Value Index mutual
fund (Value) for the years 1984 throuth 2019 provides important information for an investor trying to determine whether
to invest in a growth mutual fund, a value mutual fund, or both types of mutual funds. Over this period, the mean return
for Growth of 15.755% is greater than the mean return for Value of 12.005%. While the mean return typically represents
TZIDO SUN/Shutterstock the reward of investing, it does not incorporate the risk of investing.
Standard deviation tends to be the most common measure of risk with financial data. Since the standard deviation for
Growth (23.799%) is greater than the standard deviation for Value (17.979%), Growth is likelier to have returns farther
I N T R O D U C T O R Y C A S E above and below its mean. Finally, given a risk-free rate of 2%, the Sharpe ratio for Growth is 0.58 compared to that for
Value of 0.56, indicating that Growth provides more reward per unit of risk. Assuming that the behavior of these returns will
Investment Decision
continue, the investor will favor investing in Growth over Value. A commonly used disclaimer, however, states that past
performance is no guarantee of future results.
Dorothy Brennan works as a financial advisor at a large investment firm. She meets with an inex-
perienced investor who has some questions regarding two approaches to mutual fund investing:
growth investing versus value investing. The investor has heard that growth funds invest in com-
108 B u S I n e S S S TAT I S T I C S 3.5 Mean-Variance Analysis and the Sharpe Ratio
panies whose stock prices are expected to grow at a faster rate, relative to the overall stock mar-
ket. Value funds, on the other hand, invest in companies whose stock prices are below their true
worth. The investor has also heard that the main component of investment return is through capi-
tal appreciation in growth funds and through dividend income in value funds.
The investor shows Dorothy the annual return data for Fidelity’s Growth Index mutual fund
(Growth) and Fidelity’s Value Index mutual fund (Value). Table 3.1 shows a portion jag16309_ch03_080-123
of the annual108 07/22/20 04:00 PM
return (in %) for these two mutual funds from 1984 to 2019. It is difficult for the investor to draw
any conclusions from the data in its present form. In addition to clarifying the style differences
in growth investing versus value investing, the investor requests Dorothy to summarize
Growth_Value
the data.
TABLE 3.1 Annual Returns (in %) for Growth and Value
Year Growth Value
1984 −5.50 −8.59
1985 39.91 22.10
⋮ ⋮ ⋮
2019 38.42 31.62
Dorothy will use the sample information to:

1. Calculate and interpret the typical return for these two mutual funds.
2. Calculate and interpret the investment risk for these two mutual funds.
3. Determine which mutual fund provides the greater return relative to risk.
A synopsis of this case is provided at the end of Section 3.4.
81
jag16309_ch03_080-123 81 07/22/20 04:00 PM
x B U S I N E S S S TAT I S T I C S WALKTHROUGH
ISTUDY
and Build Skills to Communicate
Results
Writing with Data
One of our most important innovations is the inclusion of a sample report within every
chapter. Our intent is to show students how to convey statistical information in written
form to those who may not know detailed statistical methods. For example, such a report
may be needed as input for managerial decision making in sales, marketing, or company
planning. Several similar writing exercises are provided at the end of every Writing with
Data section. Each chapter also includes a synopsis that addresses questions raised from
the introductory case. This serves as a shorter writing sample for students. Instructors
Confirming Pages
of large sections may find these reports useful for incorporating writing into their
statistics courses.
Confirming Pages
6 . 4 W R I T I N G W I T H DATA
Case Study
Professor Lang is a professor of economics at Salem State university. She has been
6 . 4teaching
W R IaTcourse
I N Gin W
Principles
I T H ofD economics
ATA for over 25 years. Professor Lang has
never graded on a curve since she believes that relative grading may unduly penalize
Case Study
(benefit) a good (poor) student in an unusually strong (weak) class. She always uses an
absolute scale for making grades, as shown in the two left columns of table 6.4.
Professor Lang is a professor of economics at Salem State University. She has been
teaching a course in Principles
TABLE of Economics
6.4 Grading for over
Scales with 25 years.
Absolute GradingProfessor Lang has
versus Relative Grading
never graded on a curve since she believes that relative grading may unduly penalize
Absolute
(benefit) a good (poor) student Grading
in an unusually strong (weak) class. SheRelative Grading
always uses an
absolute scale forGrade
making grades, as shown
Scorein the two left columns
Gradeof Table 6.4. Probability
A Grading Scales with Absolute
TABLE 6.4 92 and above
Grading A Grading
versus Relative 0.10
B
Absolute Grading 78 up to 92 Relative
B Grading 0.35
Grade C Score 64 up Grade
to 78 C Probability 0.40
A D 92 and above
58 up Ato 64 D 0.10 0.10 Jasper White/image Source
B 78 up to 92 B 0.35
F Below 58 F 0.05
C 64 up to 78 C 0.40
D A colleague of58 up to 64 Lang’s D has convinced her 0.10 Image Source, all rights reserved.
Professor to move to relative grading, because it
F
corrects Below 58 problems.
for unanticipated F Professor Lang decides 0.05 to experiment with grading based
on the relative
A colleague scaleLang’s
of Professor as shown in the two
has convinced herright columns
to move of table
to relative 6.4.
grading, usingitthis relative grading
because
scheme,
corrects the top 10%
for unanticipated of students
problems. Professorwill
Langget A’s, the
decides next 35%with
to experiment B’s,grading
and sobased on. Based on her years
on theofrelative scale experience,
teaching as shown in the two right columns
Professor of Table 6.4.
Lang believes thatUsing this relative
the scores grading
in her course follow a normal
scheme, the top 10%with
distribution of students
a meanwillof get
78.6A’s,and
the a
next 35% B’s,deviation
standard and so on.of Based
12.4.on her years
of teachingProfessor
experience, Professor Lang believes that the scores in her course follow a normal
Lang wants to use this information to calculate probabilities based on the absolute
distribution with a mean of 78.6 and a standard deviation of 12.4.
scale and compare them to the probabilities based on the relative scale. then, she wants to cal-
Professor Lang wants to use this information to calculate probabilities based on the absolute
culate the range
scale and compare them to ofthe
scores for grades
probabilities basedbased
on the on the relative
relative scale
scale. Then, sheand
wantscompare
to cal- them to the abso-
culate lute scale.ofFinally,
the range she
scores for wantbased
grades to determine whichscale
on the relative grading scale makes
and compare them toit the
harder
abso-to get higher grades.
lute scale. Finally, she want to determine which grading scale makes it harder to get higher grades.
Sample
Many teachers would confess that grading is one of the most difficult tasks of their profession.
Sample
Report—
Many teachers would confess that grading is one of the most difficult tasks of their profession.
two common
Two common gradingused
grading systems systems used
in higher in higher
education are education
relative and are relative
absolute. Report—
and absolute. Relative
Relative
grading
systemssystems are norm-referenced or curve-based, in iswhich
basedaongrade
the is based on the
grading are norm-referenced
position in
or curve-based,
class. Absolute
in which a grade
grading systems, on the other Absolute
student’s relative position in class. Absolute grading systems, on the other hand, are criterion- hand, are criterion-
student’s relative Absolute
referenced,
short, short,
in whichin
referenced,
with absolute
a grade
whichisarelated
grading, the
with absolute
gradetoisthe
student’s
grading,
student’s
related
thescore
absolute
to the
is compared
student’s
performance
student’s absolute
scoretoisa compared
predetermined
in class. In
performance
to ascale,
Grading
in class. in
predetermined scale,
Grading
whereas with relative grading, the score is compared to the scores of other students in
whereas with relative grading, the score is compared to the scores of other students in versus versus
the class.
Letthe class. a grade in Professor Lang’s class, which is normally distributed with a mean Relative
X represent
Relative
of 78.6 andLet X represent
a standard a grade
deviation inThis
of 12.4. Professor Lang’s
information class,
is used whichthe
to derive is grade
normally
Grading
distributed with a mean
probabili-
ties based
P(X ≥ 92) =
on the
of 78.6
P(Z
and
≥
absolute
1.08) =
scale.deviation
a standard
0.14.
For instance,
Other
the probability
of 12.4.
probabilities, derived
of receiving
this information
similarly, are
is an
used
presented
ties based on the absolute scale. For instance, the probability of receiving
A isto
in
derived
deriveasthe grade probabili-
Table 6.5. an A is derived as
Grading
P(X ≥ 92) = P(Z ≥ 1.08) = 0.14. Other probabilities, derived similarly, are presented in table 6.5.
TABLE 6.5 Probabilities Based on Absolute Scale and Relative Scale
Probability
TABLE Based on Based
6.5 Probabilities Probability Based
on Absolute onand Relative Scale
Scale
Grade Absolute Scale Relative Scale
A Probability Based on
0.14 0.10 Probability Based on
B Grade 0.38 Absolute Scale 0.35 Relative Scale
C A 0.36 0.14 0.40 0.10
D B 0.07 0.38 0.10 0.35
F C 0.05 0.36 0.05 0.40
D 0.07 0.10
F 0.05 0.05
229
229
jag16309_ch06_200-237 229 09/01/20 08:17 PM
jag16309_ch06_200-237 229 07/12/20 01:25 PM
WALKTHROUGH
ISTUDY
Unique Coverage and
Presentation . . .
Unique Coverage of Regression Analysis
Our coverage of regression analysis is more extensive than that of the vast majority of texts.
This focus reflects the topic’s importance in the emerging field of data analytics. We combine
simple and multiple regression in one chapter, which we believe is a seamless grouping and
eliminates needless repetition. The detailed Excel and R instructions eliminate the need for
tedious manual calculations. Three more in-depth chapters cover statistical inference, nonlinear
relationships, dummy variables, and the linear probability and the logistic regression models.
The emphasis in all chapters is on conceptualization, model selection, and interpretation of
the results with reference to professionally created figures and tables.
Chapter 14: Regression Analysis
Chapter 15: Inference with Regression Models
Chapter 16: Regression Models for Nonlinear Relationships
Chapter 17: Regression Models with Dummy Variables
The authors have put forth a novel and innovative way to present
regression which in and of itself should make instructors take a long
and hard look at this book. Students should find this book very
readable and a good companion for their course.
Harvey A. Singer, George Mason University
Inclusion of Important Topics

We have incorporated several important topics, often ignored in traditional textbooks,
including data preparation, pivot tables, geometric mean return, Sharpe ratio, accuracy
rates, etc. From our experience from working outside the classroom, we have found that
professionals use these topics on a regular basis.
THE SHARPE RATIO

The Sharpe ratio measures the extra reward per unit of risk. The Sharpe ratio for an
investment I is computed as
¯f 
¯  I  −  R
x
_____
   ,
s I
¯
where x¯I  is the mean return for the investment, R
  f is the mean return for a risk-free asset
such as a Treasury bill (T-bill), and sI is the standard deviation for the investment.
Written as Taught
We introduce topics just the way we teach them; that is, the relevant tools follow the
opening application. Our roadmap for solving problems is
1. Start with intuition
2. Use Excel or R to estimate the appropriate model,
3. Communicate the results.
We use worked examples throughout the text to illustrate how to apply concepts to solve
real-world problems.
xii B U S I N E S S S TAT I S T I C S WALKTHROUGH
ISTUDY
that Make the Content More
Effective
EXAMPLE 5.9
People turn to social media to stay in touch with friends and family members, catch
the news, look for employment, and be entertained. According to a recent survey,
Integration of Microsoft
68% of all U.S. Excel
adults are Facebook and
users. R a sample of 100 randomly
Consider
selected
We prefer thatAmerican adults.
students first focus on and absorb the statistical material before replicating
their a.
results
Whatwith a computer.
is the probabilitySolving each70application
that exactly manually
American adults provides students with
are Facebook
a deeperusers?
understanding of the relevant concept. However, we recognize that embedding
computer
b. Whatoutput is probability
is the often necessary
that noinmore
order to 70
than avoid cumbersome
American calculations or the
adults are
need for Facebook
statistical users?
tables. Microsoft Excel and R are the primary software packages used
in this
c. text.
WhatWe is chose Excel and
the probability thatRatover
leastother statisticaladults
70 American packages based onusers?
are Facebook their wide-
spread use and reviewer feedback. For instructors who prefer to focus only on Excel, the
R instructions sections are easily skipped. We provide brief guidelines for using Minitab,
SPSS,SOLUTION:
and JMPWe in let
chapter appendices.
X denote the number of American adults who are Facebook users.
We also know that p = 0.68 and n = 100.
Using Excel
We use Excel’s BINOM.DIST function to calculate binomial probabilities. We
enter =BINOM.DIST(x, n, p, TRUE or FALSE) where x is the number of suc-
cesses, n is the number of trials, and p is the probability of success. For the last
argument, we enter TRUE if we want to find the cumulative probability function
P(X ≤ x) or FALSE if we want to find the probability mass function P(X = x).
a. In order to find the probability that exactly 70 American adults are Facebook
users, P(X = 70), we enter =BINOM.DIST(70, 100, 0.68, FALSE) and Excel
returns 0.0791.
b. In order to find the probability that no more than 70 American adults are
Facebook users, P(X ≤ 70), we enter =BINOM.DIST(70, 100, 0.68, TRUE)
and Excel returns 0.7007.
c. In order to find the probability that at least 70 American adults are Facebook
users, P(X ≥ 70) = 1 − P(X ≤ 69), we enter =1–BINOM.DIST(69, 100, 0.68,
TRUE) and Excel returns 0.3784.
Using R
We use R’s dbinom and pbinom functions to calculate binomial probabilities. In
order to calculate the probability mass function P(X = x), we enter dbinom(x, n, p)
where x is the number of successes, n is the number of trials, and p is the probabil-
ity of success. In order to calculate the cumulative probability function P(X ≤ x),
we enter pbinom(x, n, p).
a. In order to find P(X = 70), we enter:
> dbinom(70, 100, 0.68)
And R returns: 0.07907911.
b. In order to find P(X ≤ 70), we enter:
> pbinom(70, 100, 0.68)
c. In order to find P(X ≥ 70) = 1 − P(X ≤ 69), we enter:
> 1 − pbinom(69, 100, 0.68)
CHAPTeR 5 Discrete Probability Distributions B u S I n e S S S TAT I S T I C S 181

WALKTHROUGH
ISTUDY
Real-World Exercises and Case
Studies that Reinforce the Material
Mechanical and Applied Exercises
Chapter exercises are a well-balanced blend of mechanical, computational-type problems
followed by more ambitious, interpretive-type problems. We have found that simpler
drill problems tend to build students’ confidence prior to tackling more difficult applied
problems. Moreover, we repeatedly use many data sets—including house prices, sales,
personality types, health measures, expenditures, stock returns, salaries, and debt—in
various chapters of the text. For instance, students first use these real Rev.
dataConfirming Pages
to calculate
summary measures, make statistical inferences with confidence intervals and hypothesis
tests, and finally, perform regression analysis.
E X E RC I S E S 3.4
Mechanics 44. StockPrices. Monthly closing stock prices for Firm A
39. Consider the following population data: and Firm B are collected for the past five years. A portion of
the data is shown in the accompanying table.
34 42 12 10 22
Observation Firm A Firm B
a. Calculate the range. 1 39.91 42.04
b. Calculate MAD.
2 42.63 41.64
c. Calculate the population variance.
⋮ ⋮ ⋮
d. Calculate the population standard deviation.
60 87.51 75.09
40. Consider the following population data:
a. Calculate the sample variance and the sample standard
0 −4 2 −8 10 deviation for each firm’s stock price.
a. Calculate the range. b. Which firm’s stock price had greater variability as
measured by the standard deviation?
b. Calculate MAD.
c. Which firm’s stock price had the greater relative
dispersion?
45. Rental. A real estate analyst examines the rental mar-
41. Consider the following sample data:
ket in a college town. She gathers data on monthly rent and
40 48 32 52 38 42 the square footage for 40 apartments. A portion of the data is
shown in the accompanying table.
a. Calculate the range.
b. Calculate MAD. Monthly Rent Square Footage
c. Calculate the sample variance. 645 500
d. Calculate the sample standard deviation. 675 648
42. Consider the following sample data: ⋮ ⋮
2,400 2,700
− 10 12 −8 −2 −6 8
a. Calculate the mean and the standard deviation for
a. Calculate the range. monthly rent.
b. Calculate MAD. b. Calculate the mean and the standard deviation for square
c. Calculate the sample variance and the sample standard footage.
deviation. c. Which sample data exhibit greater relative dispersion?
46. Revenues. The accompanying data file shows the
Applications annual revenues (in $ millions) for Corporation A and Corpora-
43. Prime. The accompanying table shows a portion of the tion B for the past 13 years.
annual expenditures (in $) for 100 Prime customers. a. Calculate the coefficient of variation for Corporation A.
b. Calculate the coefficient of variation for Corporation B.
Customer Expenditures c. Which variable exhibits greater relative dispersion?
1 1272 47. Census. The accompanying data file shows, among
2 1089 other variables, median household income and median house
⋮ ⋮ value for the 50 states.
100 1389 a. Calculate and discuss the range of household income
and house value.
a. What were minimum expenditures? What were maximum b. Calculate the sample MAD and the sample standard
expenditures? deviation of household income and house value.
b. Calculate the mean and the median expenditures. c. Discuss why we cannot directly compare the sample
c. Calculte the variance and the standard devation. MAD and the standard deviations of the two variables.
xiv B U S I N E S S S TAT I S T I C S WALKTHROUGH

106 B u S I n e S S S TAT I S T I C S 3.4 Measures of Dispersion
ISTUDY
Suggested Case Studies
Here are some suggestions for analysis.
Features that Go Beyond the

Report 14.1 Perform a similar analysis to the one conducted in this section, but choose similar
mutual funds from a different investment firm, such as Fidelity or T. Rowe Price. Go to https://
finance.yahoo.com/ to extract monthly prices for each mutual fund. Calculate the monthly return
as (Pt − Pt−1)/Pt−1, were Pt is the price at time t and Pt−1 is the price at time t−1.
Typical Report 14.2 House_Price. Choose a college town. Find the model that best predicts
the sale price of a house. Use goodness-of-fit measures to find the appropriate explanatory
variables.
First Pages
Conceptual
Report 14.3
Review
enrolled students
College_Admissions. Choose a college of interest and use the sample of
to best predict a student’s college grade point average. Use goodness-of-fit
measures to find the appropriate explanatory variables. In order to estimate these models, you
At the endhave
oftoeach chapter, we present a conceptual review that provides a more holistic
first filter the data to include only the enrolled students.
approach Report
to reviewing
14.4
the material. This section revisits the learning outcomes and pro-
NBA. Develop and compare two models for predicting a player’s salary
vides the most
modelonimportant
based uses
offensive onedefinitions,
only performance
explanatory
measuresinterpretations,
variable to predict
and defensive anddescribe
and/or
performance formulas.
changes
measures, in the
respectively.
response
In order to variable. The models,
estimate these model isyouexpressed as filter
have to first β0 +data
y = the β1xto+include
ε, where and xstatistics
onlyycareer are the
response
based variable
on regular and theexclude
seasons. explanatory
playersvariable, respectively,
with no information and ε is the random error
on salary.
term. The coefficients β0 and β1 are the unknown parameters to be estimated.
We apply the ordinary least squares (OLS) method to find a sample regression equation
ŷ = b0 + b1x, where ŷ is the predicted value of the response variable and b0 and b1 are the
estimates of β0 and β1, respectively. The estimated slope coefficient b1 represents the
C O nin C
change e Px Tchanges
ŷ when U A byL oneRunit.
e VTheI eunits
Wof b1 are the same as those of y.
LO Estimate aand
14.3 Conduct
LO 14.1 interprettest
hypothesis the for
multiple linear regression
the population model.
correlation coefficient.
The multiple
When linear
conducting regression
tests concerningmodel allows
the population
____
more than
correlation one explanatory
coefficient ρ variable
xy to of
, the value be
linearly related with the response variable rxy√n y. 2 is defined as y = β0 + β1x1 + β2x2 + ⋯ +
− It
the _________
βkxktest
+ ε,statistic
where isy is
calculated as tdf variable,
the response = _____ x1, x, 2where is the
, . . ., xrkxyare the sample correlation
k explanatory coef-
variables,
and ε isand
ficient thedfrandom √1 − r 2xy β0, β1, . . ., βk are the unknown parame-
error term. The coefficients
= n − 2.
ters to be estimated. We again use the OLS method to arrive at the following sample
regression equation: ŷand
LO 14.2 Estimate 0 + b1x1 +the
= binterpret + ⋯ +linear
b2x2simple bkxk, where b0, b1, model.
regression . . ., bk are the esti-
mates of β0, β1, . . ., βk, respectively.
Regression analysis presumes that one variable, called the response variable, is influ-
For each
enced by explanatory variable
other variables, (j =explanatory
calledxj the 1, . . ., k), the corresponding
variables. The simpleslopelinear
coefficient bj is
regression
the estimated regression coefficient. It measures the change in the predicted value of the
response variable ŷ, given a unit increase in the associated explanatory variable xj, hold-
ing all other explanatory variables constant. In other words, it represents the partial
influence of xj on ŷ. CHAPTeR 14 Regression Analysis BUSIneSS S TAT I S T I C S 513
LO 14.4 Interpret goodness-of-fit measures.

The standard error of the estimate se is the standard deviation of the residual and is calcu-
______
lated as se = √ _____ SSE
n − k − 1 , where SSE is the error sum of squares. The standard error of the
jag16309_ch14_486-517 513 06/17/20 02:00 PM
estimate is a useful goodness-of-fit measure when comparing models; the model with the
smaller se provides the better fit.
The coefficient of determination R2 is the proportion of the sample variation in the
response variable that is explained by the sample regression equation. It falls between 0
and 1; the closer the value is to 1, the better the model fits the sample data.
Adjusted R2 adjusts R2 by accounting for the number of explanatory variables k used in
the regression. In comparing competing models with different numbers of explanatory
variables, the preferred model will have the highest adjusted R2.
ADDITIOnAL eXeRCISeS
59. Energy_Health. The following table shows a b. Specify the competing hypotheses in order to
portion of the monthly returns data (in percent) for determine whether the population correlation
2010–2016 for two of Vanguard’s mutual funds: the coefficient is different from zero.
Energy Fund and the Healthcare Fund. c. At the 5% significance level, what is the
conclusion to the test? Are the returns on the
Date Energy Healthcare mutual funds correlated?
Jan-10 −4.86 −0.13 60. Yields. While the Federal Reserve controls
Feb-10 1.50 0.58 short-term interest rates, long-term interest
⋮ ⋮ ⋮ rates essentially depend on supply and demand
Dec-16 −0.30 −5.26 dynamics, as well as longer-term interest rate
expectations. The accompanying table shows
a. Calculate and interpret the sample correlation a portion of data for the annualized rates for
coefficient rxy. the 10-year Treasury yield (in %) and the
514 B U S I n e S S S TAT I S T I C S Conceptual Review
jag16309_ch14_486-517 514 06/17/20 02:00 PM
WALKTHROUGH
ISTUDY
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
ISTUDY
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
ISTUDY
Remote Proctoring & Browser-Locking Capabilities
New remote proctoring and browser-locking capabilities, hosted by Proctorio within

Connect, provide control of the assessment environment by enabling security options and
verifying the identity of the student.
Seamlessly integrated within Connect, these services allow instructors to control stu-
dents’ 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.
What Resources are Available for

Instructors?
Instructor Library
The Connect Instructor Library is your repository for additional resources to improve stu-
dent engagement in and out of class. You can select and use any asset that enhances your
lecture. The Connect Instructor Library includes:
∙ PowerPoint presentations
∙ Excel Data Files
∙ Test Bank
∙ Instructor’s Solutions Manual
∙ Digital Image Library
Tegrity Campus:
Lectures 24/7
Tegrity Campus is integrated in Connect to help make your class time available 24/7.
With Tegrity, you can capture each one of your lectures in a searchable format for stu-
dents to review when they study and complete assignments using Connect. With a simple
one-click start-and-stop process, you can capture everything that is presented to students
during your lecture from your computer, including audio. Students can replay any part of
any class with easy-to-use browser-based viewing on a PC or Mac.
Educators know that the more students can see, hear, and experience class resources,
the better they learn. In fact, studies prove it. With Tegrity Campus, students quickly
recall key moments by using Tegrity Campus’s unique search feature. This search helps
students efficiently find what they need, when they need it, across an entire semester of
class recordings. Help turn all your students’ study time into learning moments immediately
supported by your lecture. To learn more about Tegrity, visit http://tegritycampus.mhhe.com.
xviii B U S I N E S S S TAT I S T I C S WALKTHROUGH
ISTUDY
ALEKS
ALEKS is an assessment and learning program that provides individualized instruction
in Business Statistics, Business Math, and Accounting. Available online in partnership
with McGraw Hill, ALEKS interacts with students much like a skilled human tutor, with
the ability to assess precisely a student’s knowledge and provide instruction on the exact
topics the student is most ready to learn. By providing topics to meet individual stu-
dents’ needs, allowing students to move between explanation and practice, correcting and
analyzing errors, and defining terms, ALEKS helps students to master course content
quickly and easily.
ALEKS also includes an instructor module with powerful, assignment-driven fea-
tures and extensive content flexibility. ALEKS simplifies course management and allows
instructors to spend less time with administrative tasks and more time directing student
learning. To learn more about ALEKS, visit www.aleks.com.
MegaStat for Microsoft Excel

MegaStat by J. B. Orris of Butler University is a full-featured Excel add-in that is
available online through the MegaStat website at www.mhhe.com/megastat or through
an access card packaged with the text. It works with Excel 2016, 2013, and 2010
(and Excel: Mac 2016). On the website, students have 10 days to successfully download
and install MegaStat on their local computer. Once installed, MegaStat will remain active in
Excel with no expiration date or time limitations. The software performs statistical analyses
within an Excel workbook. 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 its ease-of-use features include
Auto Expand for quick data selection and Auto Label detect. Since 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.
Screencam tutorials are included that provide a walkthrough of major business statistics
topics. Help files are built in, and an introductory user’s manual is also included.
WALKTHROUGH
ISTUDY
What Resources are Available for
Students?
Rev. Confirming Pages
E X E RC I S E S 3.4 Integration of Excel Data Sets. A convenient fea-

Mechanics 44. StockPrices. Monthly closing stock prices for Firm A ture is the inclusion of an Excel data file link in many
problems using data files in their calculation. The link
39. Consider the following population data: and Firm B are collected for the past five years. A portion of
the data is shown in the accompanying table.
a.
34 42
Calculate the range.

12 10 22
Observation Firm A Firm B allows students to easily launch into Excel, work the
b. Calculate MAD.
1
2
39.91
42.63
42.04
41.64
problem, and return to Connect to key in the answer
and receive feedback on their results.
⋮ ⋮ ⋮
60 87.51 75.09
40. Consider the following population data:
a. Calculate the sample variance and the sample standard
0 −4 2 −8 10 deviation for each firm’s stock price.
a. Calculate the range. b. Which firm’s stock price had greater variability as
measured by the standard deviation?
b. Calculate MAD.
c. Which firm’s stock price had the greater relative
dispersion?
45. Rental. A real estate analyst examines the rental mar-
41. Consider the following sample data:
ket in a college town. She gathers data on monthly rent and
40 48 32 52 38 42 the square footage for 40 apartments. A portion of the data is
shown in the accompanying table.
a. Calculate the range.
b. Calculate MAD. Monthly Rent Square Footage
c. Calculate the sample variance. 645 500
d. Calculate the sample standard deviation. 675 Confirming 648
Pages
42. Consider the following sample data: ⋮ ⋮
2,400 2,700
− 10 12 −8 −2 −6 8
a. Calculate the mean and the standard deviation for
a. Calculate the range. monthly rent.
b.manufacturer
The Calculatetested
MAD.12 of the cars and clocked their per- b. Calculate the mean and the standard deviation for square
51. Debt_Payments. The data accompanying this exercise
c. Calculate
formance the ofsample
times. Three the carsvariance
clocked inand theseconds,
at 5.8 footage.
sample standardshow the average debt payments (Debt, in $) for 26 metropoli-
5 cars atdeviation.
5.9 seconds, 3 cars at 6.0 seconds, and 1 car at 6.1 tan areas. c. Which sample data exhibit greater relative dispersion?
seconds. At the 5% level of significance, test if the new sports a. State46.the null and the alternativeThehypotheses in order data
to file shows the
Revenues. accompanying
Applications
car is meeting its goal to go from 0 to 60 miles per hour in less test whether average monthly debt payments are greater
annual revenues (in $ millions) for Corporation A and Corpora-
than 6 seconds. Assume a normal distribution for the analysis. than $900.
43. Prime. The accompanying table shows a portion of the tion B for the past 13 years.
48. APR. A mortgage analyst collects data from seven financial b. What assumption regarding the population is necessary
annual expenditures (in $) forrate
100(APR)
Prime
for acustomers.
institutions on the annual percentage 30-year fixed in order toa.implement
Calculatepartthe
a? coefficient of variation for Corporation A.
loan. The data accompanying this exercise show the results. c. Calculateb. Calculate
the value of the the
test coefficient of variation
statistic and the p-value. for Corporation B.
a. State the null Customer Expenditures
and the alternative hypothesis in order to d. At α = 0.05, c. areWhich
average monthlyexhibits
variable debt payments
greatergreater
relative dispersion?
test whether the mean mortgage rate for the population than $900? Explain.
1 1272 47. Census. The accompanying data file shows, among
exceeds 4.2%.
2 1089 52. Highway_Speeds. A police officer is concerned about
b. What assumption regarding the population is necessary other variables, median household income and median house
speeds on a certain section of Interstate 95. The data accom-
in order to implement⋮ part a? ⋮ value for
panying this exercise thethe50speeds
show states.of 40 cars on a Satur-
c. Calculate the value 100of the test statistic and
1389 the p-value. day afternoon.a. Calculate and discuss the range of household income
d. At a 10% significance level, what is the conclusion to the test? a. The speed limitand house
on this value.
portion of Interstate 95 is
a. Does the mean mortgage rate for the population exceed 4.2%?
What were minimum expenditures? What were maximum 65 mph. Specify the competing
b. Calculate hypotheses
the sample MAD in and
orderthe sample standard
49. PE_Ratio. A price-earnings ratio or P/E ratio is
expenditures? to determine if the average speed is greater than the
deviation of household income and house value.
calculated as a firm’s share price compared to the income or speed limit.
b. Calculate the mean and the median expenditures.
profit earned by the firm per share. Generally, a high P/E ratio c. Discuss why we cannot directly compare the sample
b. Calculate the value of the test statistic and the p-value.
c. Calculte
suggests the variance
that investors and higher
are expecting the standard
earnings devation.
growth c. At α = 0.01, are MAD and theconcerns
the officer’s standard deviations of the two variables.
warranted?
in the future compared to firms with a lower P/E ratio. The data Explain.
accompanying this exercise show of P/E ratios for 30 firms.
53. Lottery. An article found that Massachusetts residents
a. State the null and the alternative hypotheses in order to
spent an average of $860.70 on the lottery, more than three
test whether the P/E ratio of all firms differs from 15.
times the U.S. average. A researcher at a Boston think tank
b. Calculate the value of the test statistic and the p-value. believes that Massachusetts residents spend less than this
c. At α = 0.05, does the P/E ratio of all firms differ from 15? amount. He surveys 100 Massachusetts residents and asks
Explain. them about their annual expenditures on the lottery. The data
50. MPG. The data accompanying this exercise show miles accompanying this exercise show their responses.
106 B u S I n e S S S TAT I S T I C S
per gallon (MPG) for 25 ‘super-green’ cars. 3.4 Measures of Dispersion
a. Specify the competing hypotheses to test the
a. State the null and the alternative hypotheses in order to researcher’s claim.
test whether the average MPG differs from 95. b. Calculate the value of the test statistic and the p-value.
b. Calculate the value of the test statistic and the p-value. c. At the 10% significance level, do the data support the
c. At α = 0.05, can you conclude that the average MPG researcher’s claim? Explain.
differs from 95?
jag16309_ch03_080-123 106 07/22/20 04:00 PM
9.4 HYPOtHeSIS teSt FOR tHe

LO 9.5 POPuLAtIOn PROPORtIOn
Conduct a hypothesis test for Recall that the population mean μ describes a numerical variable whereas the population
the population proportion. proportion p is the essential descriptive measure for a categorical variable. The parameter
p represents the proportion of observations with a particular attribute.
As in the case for the population mean, we estimate the population proportion
_ on the
_ the sample proportion P to estimate
basis of its sample counterpart. In particular, we use
the population proportion p. Recall that although P is based on the binomial distribution,
it can be approximated by a normal distribution in large samples. This approximation is
considered valid when np ≥ 5 and n(1 − p) ≥ 5. Since p is not known, we typically test
the sample size requirement under the hypothesized value of the population proportion
p0. In most applications, the sample size is large and the normal distribution
330 B u S I n e S S S tAt I S t I C S 9.4

Guided Examples. These narrated video walk-
Hypothesis test for the Population Proportion
throughs 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
jag16309_ch09_310-343 330 08/08/20 07:11 AM
in the chapter. The video shows the steps to take to work through an exercise. Students can go through each example
multiple times if needed.
The Connect Student Resource page is the place for students to access additional resources. The Student Resource
page offers students quick access to the recommended study tools, data files, and helpful tutorials on statistical programs.
xx B U S I N E S S S TAT I S T I C S WALKTHROUGH
ISTUDY
McGraw Hill Customer Care
Contact Information
At McGraw Hill, we understand that getting the most from new technology can be chal-
lenging. That’s why our services don’t stop after you purchase our products. You can
e-mail our product specialists 24 hours a day to get product training online. Or you can
search our knowledge bank of frequently asked questions on our support website.
For customer support, call 800-331-5094 or visit www.mhhe.com/support. One of our
technical support analysts will be able to assist you in a timely fashion.
WALKTHROUGH
ISTUDY
AC K N O W L E D G M E N T S
We would like to acknowledge the following people for providing useful comments and
suggestions for past and present editions of all aspects of Business Statistics.
John Affisco Hofstra University Sangit Chatterjee Northeastern Soheila Fardanesh Towson
Mehdi Afiat College of University University
Southern Nevada Leida Chen California Carol Flannery University of
Mohammad Ahmadi University Polytechnic State University Texas–Dallas
of Tennessee–Chattanooga Anna Chernobai Syracuse Sydney Fletcher Mississippi
Sung Ahn Washington State University Gulf Coast Community
University Alan Chesen Wright State College
Mohammad Ahsanullah Rider University Andrew Flight Portland State
University Juyan Cho Colorado State University
Imam Alam University of University–Pueblo Samuel Frame Cal Poly San
Northern Iowa Alan Chow University of South Luis Obispo
Mostafa Aminzadeh Towson Alabama Priya Francisco Purdue
University Bruce Christensen Weber State University
Ardavan Asef-Vaziri California University Vickie Fry Westmoreland
State University Howard Clayton Auburn County Community College
Antenah Ayanso Brock University Ed Gallo Sinclair Community
University Robert Collins Marquette College
Scott Bailey Troy University University Glenn Gilbreath Virginia
Jayanta Bandyopadhyay Central M. Halim Dalgin Kutztown Commonwealth University
Michigan University University Robert Gillette University of
Samir Barman University of Tom Davis University of Kentucky
Oklahoma Dayton Xiaoning Gilliam Texas Tech
Douglas Barrett University of Matthew Dean University of University
North Alabama Maine Mark Gius Quinnipiac
John Beyers University of Jason Delaney University of University
Maryland Arkansas–Little Rock Malcolm Gold Saint Mary’s
Arnab Bisi Purdue University– Ferdinand DiFurio Tennessee University of Minnesota
West Lafayette Tech University Michael Gordinier Washington
Gary Black University of Matt Dobra UMUC University
Southern Indiana Luca Donno University of Deborah Gougeon University of
Randy Boan Aims Community Miami Scranton
College Joan Donohue University of Don Gren Salt Lake Community
Matthew Bognar University of South Carolina College
Iowa David Doorn University of Thomas G. Groleau Carthage
Juan Cabrera Ramapo College Minnesota College
of New Jersey James Dunne University of Babita Gupta CSU Monterey
Scott Callan Bentley University Dayton Bay
Gregory Cameron Brigham Mike Easley University of New Robert Hammond North
Young University Orleans Carolina State University
Kathleen Campbell St. Joseph’s Erick Elder University of Sheila Diann Hammon Athens
University Arkansas–Little Rock State University
Alan Cannon University of Ashraf ElHoubi Lamar Jim Han Florida Atlantic
Texas–Arlington University University
Michael Cervetti University of Roman Erenshteyn Goldey- Elizabeth Haran Salem State
Memphis Beacom College University
Samathy Chandrashekar Kathryn Ernstberger Indiana Jack Harshbarger Montreat
Salisbury University University College
Gary Huaite Chao University of Grace Esimai University of Edward Hartono University of
Pennsylvania–Kutztown Texas–Arlington Alabama–Huntsville
xxii
ISTUDY
Clifford Hawley West Virginia Carin Lightner North Carolina Robert Nauss University of
University A&T State University Missouri–St. Louis
Santhi Heejebu Cornell College Constance Lightner Fayetteville Satish Nayak University of
Paul Hong University of Toledo State University Missouri–St. Louis
Ping-Hung Hsieh Oregon State Scott Lindsey Dixie State Thang Nguyen California State
University College of Utah University–Long Beach
Marc Isaacson Augsburg College Ken Linna Auburn Mohammad Oskoorouchi
Mohammad Jamal Northern University–Montgomery California State University–
Virginia Community College Andy Litteral University of San Marcos
Robin James Harper College Richmond Barb Osyk University of Akron
Molly Jensen University of Jun Liu Georgia Southern Bhavik Pathak Indiana
Arkansas University University South Bend
Craig Johnson Brigham Young Chung-Ping Loh University of Melissa Patterson Chabot
University–Idaho North Florida College
Janine Sanders Jones University Salvador Lopez University of Scott Paulsen Illinois Central
of St. Thomas West Georgia College
Vivian Jones Bethune– John Loucks St. Edward’s James Payne Calhoun
Cookman University University Community College
Yogesh Joshi Kingsborough Cecilia Maldonado Georgia Norman Pence Metropolitan
Community College Southwestern State State College of Denver
Jerzy Kamburowski University University Dane Peterson Missouri State
of Toledo Farooq Malik University of University
Howard Kaplon Towson Southern Mississippi Joseph Petry University of
University Brent Marinan University of Illinois–Urbana/Champaign
Krishna Kasibhatla North Arizona Courtney Pham Missouri State
Carolina A&T State Ken Mayer University of University
University Nebraska–Omaha Martha Pilcher University of
Mohammad Kazemi University Bradley McDonald Northern Washington
of North Carolina–Charlotte Illinois University Cathy Poliak University of
Ken Kelley University of Notre Elaine McGivern Duquesne Wisconsin–Milwaukee
Dame University Simcha Pollack St. John’s
Lara Khansa Virginia Tech John McKenzie Babson University
Esther C. Klein St. Francis University Hamid Pourmohammadi
College Norbert Michel Nicholls State California State University–
Ronald Klimberg St. Joseph’s University Dominguez Hills
University John Miller Sam Houston State Tammy Prater Alabama State
Andrew Koch James Madison University University
University Virginia Miori St. Joseph’s Zbigniew H. Przasnyski Loyola
Subhash Kochar Portland State University Marymount University
University Prakash Mirchandani University Manying Qiu Virginia State
Brandon Koford Weber of Pittsburgh University
University Jason Molitierno Sacred Heart Troy Quast Sam Houston State
Randy Kolb St. Cloud State University University
University Elizabeth Moliski University of Michael Racer University of
Vadim Kutsyy San Jose State Texas–Austin Memphis
University Joseph Mollick Texas A&M Srikant Raghavan Lawrence
Francis Laatsch University of University–Corpus Christi Technological University
Southern Mississippi James Moran Oregon State Bharatendra Rai University of
David Larson University of University Massachusetts–Dartmouth
South Alabama Khosrow Moshirvaziri Michael Aaron Ratajczyk
John Lawrence California State California State University– Saint Mary’s University of
University–Fullerton Long Beach Minnesota
Shari Lawrence Nicholls State Tariq Mughal University of Tony Ratcliffe James Madison
University Utah University
Radu Lazar University of Patricia Mullins University of David Ravetch University of
Maryland Wisconsin–Madison California
David Leupp University of Kusum Mundra Rutgers Bruce Reinig San Diego State
Colorado–Colorado Springs University–Newark University
Carel Ligeon Auburn Anthony Narsing Macon State Darlene Riedemann Eastern
University–Montgomery College Illinois University
ACKNOWLEDGEMENTS
ISTUDY
David Roach Arkansas Tech Gary Smith Florida State Ahmad Vakil Tobin College of
University University Business
Carolyn Rochelle East Antoinette Somers Wayne State Tim Vaughan University of
Tennessee State University University Wisconsin–Eau Claire
Alfredo Romero North Ryan Songstad Augustana Raja Velu Syracuse University
Carolina A&T State College Holly Verhasselt University of
University Erland Sorensen Bentley Houston–Victoria
Ann Rothermel University of University Zhaowei Wang Citizens Bank
Akron Arun Kumar Srinivasan Rachel Webb Portland State
Jeff Rummel Emory University Indiana University–Southeast University
Deborah Rumsey The Ohio Anne-Louise Statt University of Kyle Wells Dixie State College
State University Michigan–Dearborn Alan Wheeler University of
Stephen Russell Weber State Scott Stevens James Madison Missouri–St. Louis
University University Mary Whiteside University of
William Rybolt Babson College Alicia Strandberg Temple Texas–Arlington
Fati Salimian Salisbury University University Blake Whitten University of Iowa
Fatollah Salimian Perdue Linda Sturges Suny Maritime Rick Wing San Francisco State
School of Business College University
Samuel Sarri College of Wendi Sun Rockland Trust Jan Wolcott Wichita State
Southern Nevada Bedassa Tadesse University of University
Jim Schmidt University of Minnesota Rongning Wu Baruch College
Nebraska–Lincoln Pandu Tadikamalta University John Yarber Northeast
Patrick Scholten Bentley of Pittsburgh Mississippi Community
University Roberto Duncan College
Bonnie Schroeder Ohio State Tarabay University of John C. Yi St. Joseph’s
University Wisconsin–Madison University
Sue Schou Boise State Faye Teer James Madison Kanghyun Yoon University of
University University Central Oklahoma
Pali Sen University of North Deborah Tesch Xavier Mark Zaporowski Canisius
Florida University College
Donald Sexton Columbia Patrick Thompson University of Ali Zargar San Jose State
University Florida University
Vijay Shah West Virginia Satish Thosar University of Dewit Zerom California State
University–Parkersburg Redlands University
Dmitriy Shaltayev Christopher Ricardo Tovar-Silos Lamar Eugene Zhang Midwestern
Newport University University State University
Soheil Sibdari University of Quoc Hung Tran Bridgewater Ye Zhang Indiana
Massachusetts–Dartmouth State University University–Purdue
Prodosh Simlai University of Elzbieta Trybus California University–Indianapolis
North Dakota State University–Northridge Yi Zhang California State
Harvey Singer George Mason Fan Tseng University of University–Fullerton
University Alabama–Huntsville Yulin Zhang San Jose State
Harry Sink North Carolina Silvanus Udoka North Carolina University
A&T State University A&T State University Qiang Zhen University of North
Don Skousen Salt Lake Shawn Ulrick Georgetown Florida
Community College University Wencang Zhou Baruch College
Robert Smidt California Bulent Uyar University of Zhen Zhu University of Central
Polytechnic State University Northern Iowa Oklahoma
The editorial staff of McGraw Hill are deserving of our gratitude for their guidance
throughout this project, especially Noelle Bathurst, Pat Frederickson, Ryan McAndrews,
Harper Christopher, Jamie Koch, and Matt Diamond.
We would also like to thank Jenna Eisenman, Matthew Hawke, and Megan Hawke for
their outstanding research assistance.
xxiv B U S I N E S S S TAT I S T I C S ACKNOWLEDGEMENTS
ISTUDY
BRIEF CONTENTS
PART O NE
Introduction
CHAPTER 1 Data and Data Preparation 2
PART TWO
Descriptive Statistics
CHAPTER 2 Tabular and Graphical Methods 32
CHAPER 3 Numerical Descriptive Measures 80
PART TH REE
Probability and Probability Distributions
CHAPTER 4 Introduction to Probability 124
CHAPTER 5 Discrete Probability Distributions 160
CHAPTER 6 Continuous Probability Distributions 200
PART FO UR
Basic Inference
CHAPTER 7 Sampling and Sampling Distributions 238
CHAPTER 8 Interval Estimation 278
CHAPTER 9 Hypothesis Testing 310
CHAPTER 10 Statistical Inference Concerning Two Populations 344
CHAPTER 11 Statistical Inference Concerning Variance 380
CHAPTER 12 Chi-Square Tests 408
PART FI VE
Advanced Inference
CHAPTER 13 Analysis of Variance 438
CHAPTER 14 Regression Analysis 482
CHAPTER 15 Inference with Regression Models 514
CHAPTER 16 Regression Models for Nonlinear Relationships 556
CHAPTER 17 Regression Models with Dummy Variables 588
PART S IX
Supplementary Topics
CHAPTER 18 Forecasting with Time Series Data 622
CHAPTER 19 Returns, Index Numbers, and Inflation 658
CHAPTER 20 Nonparametric Tests 682
A P PENDIXES
APPENDIX A Getting Started with R 721
APPENDIX B Tables 727
APPENDIX C Answers to Selected Even-Numbered Exercises 739
Glossary 755
Index 763
ISTUDY
CONTENTS
2.5 A Stem-and-Leaf Diagram 68

PA RT ON E 2.6 Writing with Data 70
Introduction Conceptual Review 72
Additional Exercises 73
CHAPTER 1 Appendix 2.1: Guidelines for Other Software Packages 76
DATA AND DATA PREPARATION 2 CHAPTER 3

1.1 Types of Data 4 NUMERICAL DESCRIPTIVE MEASURES 80
Sample and Population Data 4
Cross-Sectional and Time Series Data 5 3.1 Measures of Central Location 82
Structured and Unstructured Data 6 The Mean 82
Big Data 7 The Median 84
Data on the Web 9 The Mode 84
Using Excel and R to Calculate Measures of
1.2 Variables and Scales of Measurement 10 Central Location 85
The Measurement Scales 11 Note on Symmetry 88
1.3 Data Preparation 15 Subsetted Means 89
Counting and Sorting 15 The Weighted Mean 90
A Note on Handling Missing Values 20
3.2 Percentiles and Boxplots 93
Subsetting 20 A Percentile 93
A Note on Subsetting Based on Data Ranges 23 A Boxplot 94
1.4 Writing with Data 26 3.3 The Geometric Mean 97
Conceptual Review 28 The Geometric Mean Return 97
Additional Exercises 29 Arithmetic Mean versus Geometric Mean 98
The Average Growth Rate 99
3.4 Measures of Dispersion 101
PA RT TWO The Range 101
The Mean Absolute Deviation 102
Descriptive Statistics The Variance and the Standard Deviation 103
The Coefficient of Variation 105
CHAPTER 2 3.5 Mean-Variance Analysis and the Sharpe Ratio 107
3.6 Analysis of Relative Location 109
TABULAR AND GRAPHICAL METHODS 32 Chebyshev’s Theorem 109
2.1 Methods to Visualize a Categorical Variable 34
The Empirical Rule 110
A Frequency Distribution for a Categorical Variable 34 z-Scores 111
A Bar Chart 35 3.7 Measures of Association 114
A Pie Chart 35 3.8 Writing with Data 117
Cautionary Comments When Constructing or Interpreting Conceptual Review 119
Charts or Graphs 39
2.2 Methods to Visualize the Relationship Between
Appendix 3.1: Guidelines for Other Software Packages 123
Two Categorical Variables 42
A Contingency Table 42
A Stacked Column Chart 43
PA RT T H R E E
2.3 Methods to Visualize a Numerical Variable 48
A Frequency Distribution for a Numerical Variable 48 Probability and Probability Distributions
A Histogram 51
A Polygon 55
An Ogive 56 CHAPTER 4
Using Excel and R Construct a Polygon and an Ogive 57
2.4 More Data Visualization Methods 61
INTRODUCTION TO PROBABILITY 124
A Scatterplot 61 4.1 Fundamental Probability Concepts 126
A Scatterplot with a Categorical Variable 63 Events 126
A Line Chart 65 Assigning Probabilities 129
xxvi
ISTUDY
4.2 Rules of Probability 133 6.3 Other Continuous Probability Distributions 221
4.3 Contingency Tables and Probabilities 139 The Exponential Distribution 221
A Note on Independence with Empirical Probabilities 141 Using R for the Exponential Distribution 224
The Lognormal Distribution 224
4.4 The Total Probability Rule Using R for the Lognormal Distribution 226
and Bayes’ Theorem 144
The Total Probability Rule and Bayes’ Theorem 144 6.4 Writing with Data 229
Extensions of the Total Probability Rule Conceptual Review 231
and Bayes’ Theorem 146 Additional Exercises 232
4.5 Counting Rules 149 Appendix 6.1: Guidelines for Other Software
4.5 Writing with Data 151 Packages 235
Conceptual Review 154
PA RT F O U R
CHAPTER 5 Basic Inference
DISCRETE PROBABILITY
CHAPTER 7
DISTRIBUTIONS 160
5.1 Random Variables and Discrete Probability SAMPLING AND SAMPLING
Distributions 162 DISTRIBUTIONS 238
The Discrete Probability Distribution 162
7.1 Sampling 240
5.2 Expected Value, Variance, and Standard
Classic Case of a “Bad” Sample: The Literary Digest
Deviation 166
Debacle of 1936 240
Summary Measures 167
Trump’s Stunning Victory in 2016 241
Risk Neutrality and Risk Aversion 168
Sampling Methods 242
5.3 Portfolio Returns 171 Using Excel and R to Generate a Simple Random Sample 244
Properties of Random Variables 171
7.2 The Sampling Distribution of the Sample Mean 245
Summary Measures for a Portfolio 172
The Expected Value and the Standard Error
5.4 The Binomial Distribution 175 of the Sample Mean 246
Using Excel and R to Obtain Binomial Probabilities 180 Sampling from a Normal Population 247
5.5 The Poisson Distribution 183 The Central Limit Theorem 248
Using Excel and R to Obtain Poisson Probabilities 186 7.3 The Sampling Distribution of the Sample
5.6 The Hypergeometric Distribution 189 Proportion 252
Using Excel and R to Obtain Hypergeometric The Expected Value and the Standard Error
Probabilities 191 of the Sample Proportion 252
5.7 Writing with Data 193 7.4 The Finite Population Correction Factor 257
Case Study 193 7.5 Statistical Quality Control 259

Control Charts 260
Using Excel and R to Create a Control Chart 263
7.6 Writing With Data 267
Appendix 5.1: Guidelines for Other Software
Packages 198 Conceptual Review 269
Appendix 7.1: Derivation of the Mean and the
CHAPTER 6 ¯
Variance for X ¯
  and P
  273
CONTINUOUS PROBABILITY Appendix 7.2: Properties of Point Estimators 274
DISTRIBUTIONS 200 Packages 275
6.1 Continuous Random Variables and the Uniform
Distribution 202
The Continuous Uniform Distribution 202 CHAPTER 8
6.2 The Normal Distribution 206
Characteristics of the Normal Distribution 206
INTERVAL ESTIMATION 278
The Standard Normal Distribution 207 8.1 Confidence Interval For The Population Mean
Finding a Probability for a Given z Value 208 When σ Is Known 280
Finding a z Value for a Given Probability 210 Constructing a Confidence Interval for μ When σ Is
The Transformation of Normal Random Variables 212 Known 281
Using R for the Normal Distribution 216 The Width of a Confidence Interval 283
A Note on the Normal Approximation Using Excel and R to Construct a Confidence
of the Binomial Distribution 217 Interval for μ When σ Is Known 285
CONTENTS
ISTUDY
8.2 Confidence Interval For The Population Mean Hypothesis Test for μD 358
When σ Is Unknown 288 Using Excel and R for Testing Hypotheses about μD 361
The t Distribution 288 One Last Note on the Matched-Pairs Experiment 362
Summary of the tdf Distribution 289 10.3 Inference Concerning the Difference Between Two
Locating tdf Values and Probabilities 289 Proportions 366
Constructing a Confidence Interval for μ Confidence Interval for p1 − p2 366
When σ Is Unknown 291 Hypothesis Test for p1 − p2 367
Using Excel and R to Construct a Confidence
10.4 Writing with Data 372
Interval for μ When σ Is Unknown 292
8.3 Confidence Interval for the Population
Proportion 295 Additional Exercises 375
8.4 Selecting the Required Sample Size 298 Appendix 10.1: Guidelines for Other Software
Selecting n to Estimate μ 299 Packages 377
Selecting n to Estimate p 299
8.5 Writing with Data 302 CHAPTER 11
Additional Exercises 304 STATISTICAL INFERENCE
Appendix 8.1: Guidelines for Other Software CONCERNING VARIANCE 380
Packages 307 11.1 Inference Concerning
the Population Variance 382
Sampling Distribution of S2 382
CHAPTER 9 Finding χ2dfValues and Probabilities 383
Confidence Interval for the Population Variance 385
HYPOTHESIS TESTING 310 Hypothesis Test for the Population Variance 386
9.1 Introduction to Hypothesis Testing 312 Note on Calculating the p-Value for a Two-Tailed Test
The Decision to “Reject” or “Not Reject” Concerning σ2 387
the Null Hypothesis 312 Using Excel and R to Test σ2 387
Defining the Null and the Alternative Hypotheses 313 11.2 Inference Concerning the Ratio of Two Population
Type I and Type II Errors 315 Variances 391
9.2 Hypothesis Test For The Population Mean Sampling Distribution of S21∕ S22 391
When σ Is Known 318 Finding F (df 1,df 2)Values and Probabilities 392
The p-Value Approach 318 Confidence Interval for the Ratio of Two Population
Confidence Intervals and Two-Tailed Hypothesis Tests 322 Variances 394
One Last Remark 323 Hypothesis Test for the Ratio of Two Population
9.3 Hypothesis Test For The Population Mean Variances 395
When σ Is Unknown 325 Using Excel and R to Test σ21∕ σ22 397
Using Excel and R to Test μ When σ is Unknown 326 11.3 Writing with Data 401
9.4 Hypothesis Test for the Population Conceptual Review 403
Proportion 330 Additional Exercises 403
9.5 Writing with Data 334 Appendix 11.1: Guidelines for Other Software
Conceptual Review 336 Packages 405
Appendix 9.1: The Critical Value Approach 339 CHAPTER 12
Packages 342 CHI-SQUARE TESTS 408
12.1 Goodness-of-Fit Test for
a Multinomial Experiment 410
CHAPTER 10 Using R to Conduct a Goodness-of-Fit Test 414
STATISTICAL INFERENCE CONCERNING 12.2 Chi-Square Test for Independence 416

Calculating Expected Frequencies 417
TWO POPULATIONS 344 Using R to Conduct a Test for Independence 421
10.1 Inference Concerning the Difference Between Two 12.3 Chi-Square Tests for Normality 423
Means 346 The Goodness-of-Fit Test for Normality 423
Confidence Interval for μ1 − μ2 346 The Jarque-Bera Test 426
Hypothesis Test for μ1 − μ2 348
Writing with Data 429
Using Excel and R for Testing Hypotheses about μ1 − μ2 350
A Note on the Assumption of Normality 353 Conceptual Review 431
10.2 Inference Concerning Mean Differences 357 Additional Exercises 432
Recognizing a Matched-Pairs Experiment 357 Appendix 12.1: Guidelines for Other Software
Confidence Interval for μD 358 Packages 435
xxviii B U S I N E S S S TAT I S T I C S CONTENTS
ISTUDY
CHAPTER 15
PART FIVE
Advanced Inference INFERENCE WITH REGRESSION MODELS 514
15.1 Tests of Significance 516
Test of Joint Significance 516
CHAPTER 13 Test of Individual Significance 518
Using a Confidence Interval to Determine Individual
ANALYSIS OF VARIANCE 438 Significance 520
A Test for a Nonzero Slope Coefficient 521
13.1 One-Way Anova Test 440
Reporting Regression Results 523
Between-Treatments Estimate of σ2: MSTR 441
15.2 A General Test of Linear Restrictions 527
Within-Treatments Estimate of σ2: MSE 442
Using R to Conduct Partial F Tests 530
The One-Way ANOVA Table 444
Using Excel and R to Construct a One-Way ANOVA
15.3 Interval Estimates for the Response Variable 532
Using R to Find Interval Estimates for the Response
Table 444
Variable 535
13.2 Multiple Comparison Methods 449
15.4 Model Assumptions and Common Violations 537
Fisher’s Least Significant Difference (LSD) Method 449 Residual Plots 537
Tukey’s Honestly Significant Difference (HSD) Method 450 Assumption 1. 538
Using R to Construct Tukey Confidence Intervals Detecting Nonlinearities 538
for μ1 − μ2 452
Remedy 539
13.3 Two-Way Anova Test: No Interaction 456 Assumption 2. 539
The Sum of Squares for Factor A, SSA 458 Detecting Multicollinearity 540
The Sum of Squares for Factor B, SSB 459 Remedy 541
The Error Sum of Squares, SSE 459 Assumption 3. 541
Using Excel and R for a Two-Way ANOVA Test—No Detecting Changing Variability 541
Interaction 460 Remedy 542
13.4 Two-Way Anova Test: With Interaction 465 Assumption 4. 543
The Total Sum of Squares, SST 466 Detecting Correlated Observations 543
The Sum of Squares for Factor A, SSA, and the Sum of Remedy 544
Squares for Factor B, SSB 466 Assumption 5. 544
The Sum of Squares for the Interaction of Factor A and Remedy 544
Factor B, SSAB 467 Assumption 6. 545
The Error Sum of Squares, SSE 468 Summary of Regression Modeling 545
Using Excel and R for a Two-Way ANOVA Test—With Using Excel and R for Residual Plots, and R for Robust
Interaction 468 Standard Errors 545
13.5 Writing with Data 472 15.5 Writing with Data 548
Conceptual Review 474 Conceptual Review 550
Additional Exercises 475 Additional Exercises 551
Appendix 13.1: Guidelines for Other Software Packages 479 Appendix 15.1: Guidelines for Other Software Packages 553
CHAPTER 16
CHAPTER 14 REGRESSION MODELS FOR

REGRESSION ANALYSIS 482
NONLINEAR RELATIONSHIPS 556
16.1 Polynomial Regression Models 558
14.1 Hypothesis Test for the Correlation Coefficient 484
The Quadratic Regression Model 558
Testing the Correlation Coefficient ρxy 485
Using R to Estimate a Quadratic Regression Model 563
Using Excel and R to Conduct a Hypothesis Test for ρxy 485
The Cubic Regression Model 564
14.2 The Linear Regression Model 488
16.2 Regression Models with Logarithms 567
The Simple Linear Regression Model 489
A Log-Log Model 568
The Multiple Linear Regression Model 493
The Logarithmic Model 570
Using Excel and R to Estimate a Linear Regression
The Exponential Model 571
Model 494
Using R to Estimate Log-Transformed Models 575
14.3 Goodness-of-Fit Measures 500 Comparing Linear and Log-Transformed Models 575
The Standard Error of the Estimate 501 Using Excel and R to Compare Linear and
The Coefficient of Determination, R2 502 Log-Transformed Models 576
The Adjusted R2 504 A Cautionary Note Concerning Goodness-of-fit
A Cautionary Note Concerning Goodness-of-fit Measures 505 Measures 577
14.4 Writing with Data 507 16.3 Writing with Data 581
Conceptual Review 509 Conceptual Review 583
Additional Exercises 510 Additional Exercises 583
Appendix 14.1: Guidelines for Other Software Packages 512 Appendix 16.1: Guidelines for Other Software Packages 585
CONTENTS
ISTUDY
CHAPTER 17 CHAPTER19
REGRESSION MODELS WITH RETURNS, INDEX NUMBERS,

DUMMY VARIABLES 588 AND INFLATION 658
17.1 Dummy Variables 590 19.1 Investment Return 660
A Categorical Explanatory Variable with Two Categories 590 The Adjusted Closing Price 661
Using Excel and R to Make Dummy Variables 592 Nominal versus Real Rates of Return 662
Assessing Dummy Variable Models 592 19.2 Index Numbers 664
A Categorical Explanatory Variable with Multiple A Simple Price Index 664
Categories 593 An Unweighted Aggregate Price Index 666
17.2 Interactions with Dummy Variables 599 A Weighted Aggregate Price Index 667
Using R to Estimate a Regression Model with a 19.3 Using Price Indices to Deflate a Time Series 672
Dummy Variable and an Interaction Variable 602 Inflation Rate 674
17.3 The Linear Probability Model and the Logistic
Regression Models 605
The Linear Probability Model 605 Conceptual Review 678
The Logistic Regression Model 606 Additional Exercises 679
Using R to Estimate a Logistic Regression Model 609
Accuracy of Binary Choice Models 609 CHAPTER 20
Using R to Find the Accuracy Rate 611
17.4 Writing with Data 613 NONPARAMETRIC TESTS 682
Conceptual Review 616 20.1 Testing a Population Median 684

Additional Exercises 617 The Wilcoxon Signed-Rank Test for a Population
Median 684
Appendix 17.1: Guidelines for Other Software Packages 620 Using a Normal Distribution Approximation for T 687
Using R to Test a Population Median 688
PA RT S IX 20.2 Testing Two Population Medians 690
The Wilcoxon Signed-Rank Test for a Matched-Pairs
Supplementary Topics Sample 690
Using R to Test for Median Differences from a Matched-
CHAPTER 18 Pairs Sample 691
The Wilcoxon Rank-Sum Test for Independent Samples 691
FORECASTING WITH TIME SERIES DATA 622 Using R to Test for Median Differences from
Independent Samples 694
18.1 The Forecasting Process for Time Series 624
Using a Normal Distribution Approximation for W 694
Forecasting Methods 625
Model Selection Criteria 625 20.3 Testing Three or More Population Medians 697
18.2 Simple Smoothing Techniques 626 The Kruskal-Wallis Test for Population Medians 697
The Moving Average Technique 627 Using R to Conduct a Kruskal-Wallis Test 699
The Simple Exponential Smoothing Technique 629 20.4 The Spearman Rank Correlation Test 700
Using R for Exponential Smoothing 631 Using R to Conduct the Spearman Rank Correlation
18.3 Linear Regression Models for Trend and Test 702
Seasonality 633 Summary of Parametric and Nonparametric Tests 703
The Linear Trend Model 633 20.5 The Sign Test 705
The Linear Trend Model with Seasonality 635
20.6 Tests Based on Runs 709
Estimating a Linear Trend Model with Seasonality
The Method of Runs Above and Below the Median 710
with R 637
Using R to Conduct the Runs Test 711
A Note on Causal Models for Forecasting 637
18.4 Nonlinear Regression Models for Trend and 20.7 Writing with Data 713
Seasonality 639 Conceptual Review 715
The Exponential Trend Model 639 Additional Exercises 717
Using R to Forecast with an Exponential Trend Model 641
The Polynomial Trend Model 642
Packages 718
Nonlinear Trend Models with Seasonality 643
Using R to Forecast a Quadratic Trend Model with
Seasons 645 APPENDIXES
18.5 Causal Forecasting Methods 647 APPENDIX A etting Started with R 721
G
Lagged Regression Models 648
APPENDIX B Tables 727
Using R to Estimate Lagged Regression Models 650
APPENDIX C Answers to Selected Even-
Numbered Exercises 739
Additional Exercises 655 Glossary 755
Appendix 18.1: Guidelines for Other Software Packages 656 Index 763
xxx B U S I N E S S S TAT I S T I C S CONTENTS
ISTUDY
BUSINESS STATISTICS
ISTUDY
1 Data and Data
Preparation
L E A R N I N G O B J E C T I V E S
After reading this chapter, you should be able to:
LO 1.1 Explain the various data types.
LO 1.2 Describe variables and types of measurement scales.
LO 1.3 Inspect and explore data.
LO 1.4 Apply data subsetting.
I
n just about any contemporary human activity, we use statistics to analyze large amounts of
data for making better decisions. Managers, consumers, sports enthusiasts, politicians, and
medical professionals are increasingly turning to data to boost a company’s revenue, deepen
customer engagement, find better options on consumer products, prevent threats and fraud, suc-
ceed in sports and elections, provide better diagnoses and cures for diseases, and so on. In this
chapter, we will describe various types of data and measurement scales of variables that are used
in statistics.
It is important to note that after obtaining relevant data, we often spend a considerable amount
of time on inspecting and preparing the data for subsequent analysis. In this chapter, we will discuss
a few important data preparation tasks. We will use counting and sorting of relevant variables to
inspect and explore data. Finally, we will discuss a commonly used technique called subsetting,
where only a portion (subset) of the data is used for the analysis.
ISTUDY
Yuliia Mazurkevych/Shutterstoick
I N T R O D U C T O R Y C A S E
Gaining Insights into Retail Customer Data

Organic Food Superstore is an online grocery store that specializes in providing organic food
products to health-conscious consumers. The company offers a membership-based service that
ships fresh ingredients for a wide range of chef-designed meals to its members’ homes. Cathe-
rine Hill is a marketing manager at Organic Food Superstore. She has been assigned to market
the company’s new line of Asian-inspired meals. Research has shown that the most likely custom-
ers for healthy ethnic cuisines are college-educated millennials (born between 1982 and 2000).
In order to spend the company’s marketing dollars efficiently, Catherine wants to focus on this
target demographic when designing the marketing campaign. With the help of the information tech-
nology (IT) group, Catherine has acquired a representative sample that includes each customer’s
identification number (CustID), sex (Sex), race (Race), birthdate (BirthDate), whether the customer
has a college degree (College), household size (HHSize), annual income (Income), total spending
(Spending), total number of orders during the past 24 months (Orders), and the channel through
which the customer was originally acquired (Channel). Table 1.1 shows a portion of the data set.
TABLE 1.1 A Sample of Organic Food Superstore Customers

CustID Sex Race BirthDate ... Channel
Customers
1530016 Female Black 12/16/1986 ... SM
1531136 Male White 5/9/1993 ... TV
⋮ ⋮ ⋮ ⋮ ⋮ ⋮
1579979 Male White 7/5/1999 ... SM
Catherine wants to use the Customers data set to:

1. Identify Organic Food Superstore’s college-educated millennial customers.
2. Compare the profiles of female and male college-educated millennial customers.
A synopsis of this case is provided at the end of Section 1.3.
ISTUDY
LO 1.1 1.1 TYPES OF DATA
Explain the various types In general, data are compilations of facts, figures, or other contents, both numerical and
of data. nonnumerical. Data of all types and formats are generated from multiple sources. Insights
from all of these data improves a company’s bottom-line and enhances consumer experi-
ence. At the core, business statistics benefits companies by developing better marketing
strategies, deepening customer engagement, enhancing efficiency in procurement, uncov-
ering ways to reduce expenses, identifying emerging market trends, mitigating risk and
fraud, etc. We often find a large amount of data at our disposal. However, we also derive
insights from relatively small data sets, such as from consumer focus groups, marketing
surveys, or reports from government agencies.
Every day, consumers and businesses use data from various sources to help make deci-
sions. In order to make intelligent decisions in a world full of uncertainty, we have to under-
stand statistics—the language of data. In the broadest sense, statistics is the science of
extracting useful information from data. Three steps are essential for doing good statistics.
An important first step for making decisions is to find the right data, which are both com-
plete and lacking any misrepresentation, and prepare it for the analysis. Second, we must
use the appropriate statistical tools, depending on the data at hand. Finally, an important
ingredient of a well-executed statistical analysis is to clearly communicate information into
verbal and written language. It is important to note that numerical results are not very useful
unless they are accompanied with clearly stated actionable business insights.
DATA AND STATISTICS

Data are compilations of facts, figures, or other contents, both numerical and non-
numerical. Statistics is the science that deals with the collection, preparation, anal-
ysis, interpretation, and presentation of data.
In the introductory case, Catherine wants to target college-educated millennials when

designing the marketing campaign so that she spends the company’s marketing dollars
efficiently. Before we analyze the information that Catherine has gathered, it is important
to understand different types of data and measurement scales of variables. In this section,
we focus on data types.
LO 1.1 Sample and Population Data

Explain the various types of We generally divide the study of statistics into two branches: descriptive statistics and
data. inferential statistics. Descriptive statistics refers to the summary of important aspects of a
data set. This includes collecting data, organizing the data, and then presenting the data in
the form of charts and tables. In addition, we often calculate numerical measures that
summarize the data by providing, for example, the typical value and the variability of the
variable of interest. Today, the techniques encountered in descriptive statistics account for
the most visible application of statistics—the abundance of quantitative information that is
collected and published in our society every day. The unemployment rate, the president’s
approval rating, the Dow Jones Industrial Average, batting averages, the crime rate, and the
divorce rate are but a few of the many “statistics” that can be found in a reputable newspaper
on a frequent, if not daily, basis. Yet, despite the familiarity of descriptive statistics, these
methods represent only a minor portion of the body of statistical applications.
The phenomenal growth in statistics is mainly in the field called inferential statistics.
Generally, inferential statistics refers to drawing conclusions about a large set of data—
called a population—based on a smaller set of sample data. A population is defined as
all members of a specified group (not necessarily people), whereas a sample is a subset of
that particular population. In most statistical applications, we must rely on sample data in
order to make inferences about various characteristics of the population.
4 B u si n e ss S tatistics 1.1 Types of Data
ISTUDY
Figure 1.1 depicts the flow of information between a population and a sample. Con-
sider, for example, a 2016 Gallop survey that found that only 50% of millennials plan to
stay at their current job for more than a year. Researchers use this sample result, called a
sample statistic, in an attempt to estimate the corresponding unknown population
parameter. In this case, the parameter of interest is the percentage of all millennials who
plan to be with their current job for more than a year.
FIGURE 1.1 Population versus Sample
Sampling
Population Sample
Inference
POPULATION VERSUS SAMPLE

A population consists of all items of interest in a statistical problem. A sample is a
subset of the population. We analyze sample data and calculate a sample statistic to
make inferences about the unknown population parameter.
It is generally not feasible to obtain population data due to prohibitive costs and/or
practicality. We rely on sampling because we are unable to use population data for two
main reasons.
∙ Obtaining information on the entire population is expensive. Consider how the
monthly unemployment rate in the United States is calculated by the Bureau of Labor
Statistics (BLS). Is it reasonable to assume that the BLS counts every unemployed
person each month? The answer is a resounding NO! In order to do this, every home in
the country would have to be contacted. Given that there are approximately 160 million
individuals in the labor force, not only would this process cost too much, it would take an
inordinate amount of time. Instead, the BLS conducts a monthly sample survey of about
60,000 households to measure the extent of unemployment in the United States.
∙ It is impossible to examine every member of the population. Suppose we are
interested in the average length of life of a Duracell AAA battery. If we tested the
duration of each Duracell AAA battery, then in the end, all batteries would be dead
and the answer to the original question would be useless.
Cross-Sectional and Time Series Data

Sample data are generally collected in one of two ways. Cross-sectional data refer to
data collected by recording a characteristic of many subjects at the same point in time, or
without regard to differences in time. Subjects might include individuals, households,
firms, industries, regions, and countries.
Table 1.2 is an example of a cross-sectional data set. It lists the team standings for the
National Basketball Association’s Eastern Conference at the end of the 2018–2019 season.
The eight teams may not have ended the season precisely on the same day and time, but the
differences in time are of no relevance in this example. Other examples of cross-sectional
data include the recorded grades of students in a class, the sale prices of single-family
homes sold last month, the current price of gasoline in different cities in the United States,
and the starting salaries of recent business graduates from the University of Connecticut.
Chapter 1 Data and Data Preparation
ISTUDY
TABLE 1.2 2018–2019 NBA Eastern Conference
Team name Wins Losses Winning percentage
Milwaukee Bucks 60 22 0.732
Toronto Raptors* 58 24 0.707
Philadephia 76ers 51 31 0.622
Boston Celtics 49 33 0.598
Indiana Pacers 48 34 0.585
Brooklyn Nets 42 40 0.512
Orlando Magic 42 40 0.512
Detroit Pistons 41 41 0.500
*The Toronto Raptors won their first NBA title during the 2018–2019 season.
Time series data refer to data collected over several time periods focusing on certain
groups of people, specific events, or objects. Time series data can include hourly, daily,
weekly, monthly, quarterly, or annual observations. Examples of time series data include
the hourly body temperature of a patient in a hospital’s intensive care unit, the daily price
of General Electric stock in the first quarter of 2020, the weekly exchange rate between
the U.S. dollar and the euro over the past six months, the monthly sales of cars at a dealer-
ship in 2020, and the annual population growth rate of India in the last decade. In these
examples, temporal ordering is relevant and meaningful.
Figure 1.2 shows a plot of the national homeownership rate in the U.S. from 2000 to
2018. According to the U.S. Census Bureau, the national homeownership rate in the first
quarter of 2016 plummeted to 63.6% from a high of 69.4% in 2004. An explanation for the
decline in the homeownership rate is the stricter lending practices caused by the housing
market crash in 2007 that precipitated a banking crisis and deep recession. This decline
can also be attributed to home prices outpacing wages in the sample period.
FIGURE 1.2 70.0

Homeownership Rate (in %)
in the U.S. from 2000 through
2018
Homeownership rate (in %)
68.0
66.0
64.0
62.0
2000 2002 2004 2006 2008 2010 2012 2014 2016 2018
Year
Structured and Unstructured Data

When you think of data, the first image that probably pops in your head is lots of numbers
and perhaps some charts and graphs. In reality, data can come in multiple forms.
For example, information exchange in social networking websites such as Facebook,
LinkedIn, and Twitter also constitute data. In order to better understand the various forms
of data, we make a distinction between structured and unstructured data.
Generally, structured data reside in a predefined, row-column format. We use
spreadsheet or database applications to enter, store, query, and analyze structured data.
ISTUDY
Examples of structured data include numbers, dates, and groups of words and numbers,
typically stored in a tabular format. Structured data often consist of numerical informa-
tion that is objective and is not open to interpretation.
Point-of-sale and financial data are examples of structured data and are usually
designed to capture a business process or transaction. Examples include the sale of retail
products, money transfer between bank accounts, and the student enrollment in a univer-
sity course. When individual consumers buy products from a retail store, each transaction
is captured into a record of structured data.
Consider the sales invoice shown in Figure 1.3. Whenever a customer places an order
like this, there is a predefined set of data to be collected, such as the transaction date,
shipping address, and the units of product being purchased. Even though a receipt or an
invoice may not always be presented in rows and columns, the predefined structure
allows businesses and organizations to translate the data on the document into a
row-column format.
FIGURE 1.3 A sample

invoice from a retail
transaction
For decades, companies and organizations relied mostly on structured data to run
their businesses and operations. Today, with the advent of the digital age, most experts
agree that only about 20% of all data used in business decisions are structured data. The
remaining 80% are unstructured.
Unlike structured data, unstructured data (or unmodeled data) do not conform to a
predefined, row-column format. They tend to be textual (e.g., written reports, e-mail mes-
sages, doctor’s notes, or open-ended survey responses) or have multimedia contents (e.g.,
photographs, videos, and audio data). Even though these data may have some implied
structure (e.g., a report title, e-mail’s subject line, or a time stamp on a photograph), they
are still considered unstructured as they do not conform to a row-column model required
in most database systems. Social media data such as Twitter, YouTube, Facebook, and
blogs are examples of unstructured data.
Big Data
Nowadays, businesses and organizations generate and gather more and more data at an
increasing pace. The term big data is a catch-phrase, meaning a massive amount of both
structured and unstructured data that are extremely difficult to manage, process, and
ISTUDY
analyze using traditional data-processing tools. Despite the challenges, big data present
great opportunities to gain knowledge and business intelligence with potential game-
changing impacts on company revenues, competitive advantage, and organizational
efficiency.
More formally, a widely accepted definition of big data is “high-volume, high-velocity
and/or high-variety information assets that demand cost-effective, innovative forms of
information processing that enable enhanced insight, decision making, and process
automation” (www.gartner.com). The three characteristics (the three Vs) of big data are:
∙ Volume: An immense amount of data is compiled from a single source or a wide
range of sources, including business transactions, household and personal devices,
manufacturing equipment, social media, and other online portals.
∙ Velocity: In addition to volume, data from a variety of sources get generated at a rapid
speed. Managing these data streams can become a critical issue for many organizations.
∙ Variety: Data also come in all types, forms, and granularity, both structured and
unstructured. These data may include numbers, text, and figures as well as audio,
video, e-mails, and other multimedia elements.
In addition to the three defining characteristics of big data, we also need to pay close
attention to the veracity of the data and the business value that they can generate. Veracity
refers to the credibility and quality of data. One must verify the reliability and accuracy
of the data content prior to relying on the data to make decisions. This becomes
increasingly challenging with the rapid growth of data volume fueled by social media and
automatic data collection. Value derived from big data is perhaps the most important
aspect of any statistical project. Having a plethora of data does not guarantee that useful
insights or measurable improvements will be generated. Organizations must develop a
methodical plan for formulating business questions, curating the right data, and unlocking
the hidden potential in big data.
Big data, however, do not necessarily imply complete (population) data. Take, for
example, the analysis of all Facebook users. It certainly involves big data, but if we
consider all Internet users in the world, Facebook users are only a very large sample.
There are many Internet users who do not use Facebook, so the data on Facebook do not
represent the population. Even if we define the population as pertaining to those who use
online social media, Facebook is still one of many social media portals that consumers
use. And because different social media are used for different purposes, data collected
from these sites may very well reflect different populations of Internet users; this
distinction is especially important from a strategic business standpoint. Therefore,
Facebook data are simply a very large sample.
In addition, we may choose not to use big data in its entirety even when they are avail-
able. Sometimes it is just inconvenient to analyze a very large data set as it is computa-
tionally burdensome, even with a modern, high-capacity computer system. Other times,
the additional benefits of working with big data may not justify the associated costs. In
sum, we often choose to work with relatively smaller data sets drawn from big data.
STRUCTURED, UNSTRUCTURED, AND BIG DATA

Structured data are data that reside in a predefined, row-column format, while
unstructured data do not conform to a predefined, row-column format. Big data is a
term used to describe a massive amount of both structured and unstructured data that
are extremely difficult to manage, process, and analyze using traditional data-processing
tools. Big data, however, do not necessary imply complete (population) data.
In this textbook, we will focus on traditional statistical methods applied to structured

data. Sophisticated tools to analyze unstructured data are beyond the scope of this
textbook.
ISTUDY
Data on the Web
The explosion in the field of statistics and data analytics is partly due to the
growing availability of vast amounts of data and improved computational power.
Many experts believe that 90% of the data in the world today were created in the
last two years alone. These days, it has become easy to access data by simply
using a search engine like Google. These search engines direct us to data-
providing websites. For instance, searching for economic data may lead you to
the Bureau of Economic Analysis (www.bea.gov), the Bureau of Labor Statistics
(www.bls.gov/data), the Federal Reserve Economic Data (research.stlouisfed.
org), and the U.S. Census Bureau (www.census.gov/data.html). These websites
provide data on inflation, unemployment, GDP, and much more, including useful
international data.
The National Climatic Data Center (www.ncdc.noaa.gov/data-access) pro-
vides a large collection of environmental, meteorological, and climate data. Sim-
ilarly, transportation data can be found at www.its-rde.net. The University of
Michigan has compiled sentiment data found at www.sca.isr.umich.edu. Several
cities in the United States have publicly available data in categories such as
finance, community and economic development, education, and crime. For
example, the Chicago data portal data.cityofchicago.org provides a large volume Comstock Images/Jupiterimages
of city-specific data. Excellent world development indicator data are available at
data.worldbank.org. The happiness index data for most countries are available at www.
happyplanetindex.org/data.
Private corporations also make data available on their websites. For example, Yahoo
Finance (www.finance.yahoo.com) and Google Finance (www.google.com/finance) list
data such as stock prices, mutual fund performance, and international market data. Zillow
(www.zillow.com/) supplies data for recent home sales, monthly rent, mortgage rates, and
so forth. Similarly, www.espn.go.com offers comprehensive sports data on both profes-
sional and college teams. Finally, The Wall Street Journal, The New York Times, USA
Today, The Economist, Business Week, Forbes, and Fortune are all reputable publications
that provide all sorts of data. We would like to point out that all of the above data sources
represent only a fraction of publicly available data.
E X E RC I S E S 1.1
Applications computes the average GPA as 3.29. Describe the relevant
population.
1. A few years ago, it came as a surprise when Apple’s iPhone
4 was found to have a problem. Users complained of weak 4. Recent college graduates with an engineering degree continue
reception, and sometimes even dropped calls, when they to earn high salaries. An online search revealed that the average
cradled the phone in their hands in a particular way. A survey annual salary for an entry-level position in engineering is $65,000.
at a local store found that 2% of iPhone 4 users experienced a. What is the relevant population?
this reception problem. b. Do you think the average salary of $65,000 is computed
a. Describe the relevant population. from the population? Explain.
b. Is 2% associated with the population or the sample? 5. Research suggests that depression significantly increases the
2. Many people regard video games as an obsession for young- risk of developing dementia later in life. Suppose that in a
sters, but, in fact, the average age of a video game player is study involving 949 elderly persons, it was found that 22% of
35 years old. Is the value 35 likely the actual or the estimated those who had depression went on to develop dementia, com-
average age of the population? Explain. pared to only 17% of those who did not have depression.
3. An accounting professor wants to know the average GPA of a. Describe the relevant population and the sample.
the students enrolled in her class. She looks up information b. Are the numbers 22% and 17% associated with the
on Blackboard about the students enrolled in her class and population or a sample?
ISTUDY
6. Go to www.zillow.com and find the sale price of 20 single-family 9. Conduct an online search to compare small hybrid vehicles
homes sold in Las Vegas, Nevada, in the last 30 days. Structure (e.g., Toyota Prius, Ford Fusion, Chevrolet Volt) on price, fuel
the data in a tabular format and include the sale price, the num- economy, and other specifications. Do you consider the search
ber of bedrooms, the square footage, and the age of the house. results structured or unstructured data? Explain.
Do these data represent cross-sectional or time series data? 10. Find Under Armour’s annual revenue from the past 10 years.
7. Go to www.finance.yahoo.com to get the current stock quote Are the data considered structured or unstructured? Explain.
for Home Depot (ticker symbol = HD). Use the ticker symbol to Are they cross-sectional or time series data?
search for historical prices and create a table that includes the 11. Ask 20 of your friends about their online social media usage,
monthly adjusted close price of Home Depot stock for the last specifically whether or not they use Facebook, Instagram,
12 months. Do these data represent cross-sectional or time and Snapchat; how often they use each social media portal;
series data? and their overall satisfaction of each of these portals. Create
8. Go to The New York Times website at www.nytimes.com and a table that presents this information. Are the data considered
review the front page. Would you consider the data on the structured or unstructured? Are they cross-sectional or time
page to be structured or unstructured? Explain. series data?
LO 1.2 1.2 VARIABLES AND SCALES OF MEASUREMENT

Describe variables and types For any statistical analysis, we invariably focus on people, firms, or events with particular
of measurement scales. characteristics. When a characteristic of interest differs in kind or degree among various
observations (records), then the characteristic can be termed a variable. Marital status
and income are examples of variables because a person’s marital status and income vary
from person to person. Variables are further classified as either categorical (qualitative)
or numerical (quantitative). The observations of a categorical variable represent catego-
ries, whereas the observations of a numerical variable represent meaningful numbers. For
example, marital status is a categorical variable, whereas income is a numerical
variable.
For a categorical variable, we use labels or names to identify the distinguishing
characteristic of each observation. For instance, a university may identify each stu-
dent’s status as either at the undergraduate or the graduate level, where the education
level is a categorical variable representing two categories. Categorical variables can
also be defined by more than two categories. Examples include marital status (single,
married, widowed, divorced, separated), IT firm (hardware, software, cloud), and
course grade (A, B, C, D, F). It is important to note that categories are often converted
into numerical codes for purposes of data processing, which we will discuss in
later chapters.
For a numerical variable, we use numbers to identify the distinguishing characteristic
of each observation. Numerical variables, in turn, are either discrete or continuous. A
discrete variable assumes a countable number of values. Consider the number of chil-
dren in a family or the number of points scored in a basketball game. We may observe
values such as 3 children in a family or 90 points being scored in a basketball game, but
we will not observe fractions such as 1.3127 children or 92.4724 scored points. The val-
ues that a discrete variable assumes need not be whole numbers. For example, the price
of a stock for a particular firm is a discrete variable. The stock price may take on a value
of $20.37 or $20.38, but it cannot take on a value between these two points.
A continuous variable is characterized by uncountable values within an interval.
Weight, height, time, and investment return are all examples of continuous variables. For
example, an unlimited number of values occur between the weights of 100 and 101
pounds, such as 100.3, 100.625, 100.8342, and so on. In practice, however, continuous
variables are often measured in discrete values. We may report a newborn’s weight (a
continuous variable) in discrete terms as 6 pounds 10 ounces and another newborn’s
weight in similar discrete terms as 6 pounds 11 ounces.
10 B u si n e ss S tatistics 1.2 Variables and Scales of Measurement
ISTUDY
CATEGORICAL AN D NUMERICAL VARIABLES
A variable is a general characteristic being observed on a set of people, objects, or
events, where each observation varies in kind or degree.
∙ The observations of a categorical variable assume names or labels.
∙ The observations of a numerical variable assume meaningful numerical val-
ues. A numerical variable can be further categorized as either discrete or con-
tinuous. A discrete variable assumes a countable number of values, whereas a
continuous variable is characterized by uncountable values within an interval.
EXAMPLE 1.1
In the introductory case, Catherine Hill has been assigned to help market Organic
Food Superstore’s new line of Asian-inspired meals. With the help of the IT group,
she has acquired a representative sample of customers at her store. Determine
which of the variables in the sample are categorical or numerical and, if numerical,
determine if they are discrete or continuous.
SOLUTION:
The variables Sex, Race, College, and Channel are categorical, merely representing
labels. We also treat Birthdate as a categorical variable, with numerous categories,
even though it contains numbers. Note that we can easily convert date of birth to a
numerical variable age by simply subtracting it from the current date. On the other
hand, HHSize, Income, Spending, and Orders are numerical variables because the
observations are all meaningful numbers. Note that all of the numerical variables in
this example are discrete because they can only assume a countable number of values;
in other words, they are not characterized by uncountable values within an interval.
The Measurement Scales

In order to choose the appropriate techniques for summarizing and analyzing variables,
we need to distinguish between the different measurement scales. The observations for
any variable can be classified into one of four major measurement scales: nominal, ordi-
nal, interval, or ratio. Nominal and ordinal scales are used for categorical variables,
whereas interval and ratio scales are used for numerical variables. We discuss these scales
in ascending order of sophistication.
The nominal scale represents the least sophisticated level of measurement. If we are
presented with nominal observations, all we can do is categorize or group them. The
observations differ merely by name or label. Table 1.3 lists the 30 publicly-owned companies,
as of February 2019, that comprise the Dow Jones Industrial Average (DJIA). The DJIA is a
stock market index that shows how these large U.S.-based companies have traded during a
standard trading session in the stock market. Table 1.3 also indicates where stocks of these
companies are traded: on either the National Association of Securities Dealers Automated
Quotations (Nasdaq) or the New York Stock Exchange (NYSE). These observations are
classified as nominal scale because we are simply able to group or categorize them. Specifically,
only five stocks are traded on the Nasdaq, whereas the remaining 25 are traded on the NYSE.
Often, we substitute numbers for the particular categorical characteristic or trait that
we are grouping. For instance, we might use the number 0 to show that a company’s stock
is traded on the Nasdaq and the number 1 to show that a company’s stock is traded on the
NYSE. One reason why we do this is for ease of exposition; always referring to the
National Association of Securities Dealers Automated Quotations, or even the Nasdaq,
can be awkward and unwieldy.
ISTUDY
TABLE 1.3 Companies of the DJIA and Exchange Where Stock is Traded
Company Exchange Company Exchange
3M (MMM) NYSE Johnson & Johnson (JNJ) NYSE
American Express (AXP) NYSE JPMorgan Chase (JPM) NYSE
Apple (AAPL) Nasdaq McDonald’s (MCD) NYSE
Boeing (BA) NYSE Merck (MRK) NYSE
Caterpillar (CAT) NYSE Microsoft (MFST) Nasdaq
Chevron (CVX) NYSE Nike (NKE) NYSE
Cisco (CSCO) Nasdaq Pfizer (PFE) NYSE
Coca-Cola (KO) NYSE Procter & Gamble (PG) NYSE
Disney (DIS) NYSE Travelers (TRV) NYSE
DowDupont (DWDP) NYSE United Health (UNH) NYSE
ExxonMobil (XOM) NYSE United Technologies (UTX) NYSE
Goldman Sachs (GS) NYSE Verizon (VZ) NYSE
Home Depot (HD) NYSE Visa (V) NYSE
IBM (IBM) NYSE Wal-Mart (WMT) NYSE
Intel (INTC) Nasdaq Walgreen (WBA) Nasdaq
Compared to the nominal scale, the ordinal scale reflects a stronger level of measure-
ment. With ordinal observations, we are able to both categorize and rank them with
respect to some characteristic or trait. The weakness with ordinal observations is that we
cannot interpret the difference between the ranked observations because the actual num-
bers used are arbitrary. Consider, for example, hotel reviews where consumers are asked
to classify the service at a particular hotel as excellent (5 stars), very good (4 stars), good
(3 stars), fair (2 stars), or poor (1 star). We summarize the categories and their respective
ratings in Table 1.4.
TABLE 1.4 Hotel Survey
Categories with Ratings
Category Rating
Excellent 5
Very good 4
Good 3
Fair 2
Poor 1
In Table 1.4, the number attached to excellent (5 stars) is higher than the number attached
to good (3 stars), indicating that the response of excellent is preferred to good. However,
we can easily redefine the ratings, as we show in Table 1.5.
TABLE 1.5 Hotel Survey

Categories with Redefined Ratings
Category Rating
Excellent 100
Very good 80
Good 70
Fair 50
Poor 40
ISTUDY
In Table 1.5, excellent still receives a higher number than good, but now the difference
between the two categories is 30 points (100 – 70), as compared to a difference of 2
points (5 – 3) when we use the first classification. In other words, differences between
categories are meaningless with ordinal observations. (We also should note that we could
reverse the ordering so that, for instance, excellent equals 40 and poor equals 100; this
renumbering would not change the nature of the observations.)
As mentioned earlier, observations of a categorical variable are typically expressed in
words but are coded into numbers for purposes of data processing. When summarizing
the results of a categorical variable, we typically count the number of observations that
fall into each category or calculate the percentage of observations that fall into each cat-
egory. However, with a categorical variable, we are unable to perform meaningful arith-
metic operations, such as addition and subtraction.
With observations that are measured on the interval scale, we are able to categorize
and rank them as well as find meaningful differences between them. The Fahrenheit scale
for temperatures is an example of interval-scaled variable. Not only is 60 degrees Fahren-
heit hotter than 50 degrees Fahrenheit, the same difference of 10 degrees also exists
between 90 and 80 degrees Fahrenheit.
The main drawback of an interval-scaled variable is that the value of zero is
arbitrarily chosen; the zero point of an interval-scaled variable does not reflect a
complete absence of what is being measured. No specific meaning is attached to
0 degrees Fahrenheit other than to say it is 10 degrees colder than 10 degrees Fahrenheit.
With an arbitrary zero point, meaningful ratios cannot be constructed. For instance, it
is senseless to say that 80 degrees is twice as hot as 40 degrees; in other words, the ratio
80/40 has no meaning.
The ratio scale represents the strongest level of measurement. The ratio scale has all
the characteristics of the interval scale as well as a true zero point, which allows us to
interpret the ratios between observations. The ratio scale is used in many business appli-
cations. Variables such as sales, profits, and inventory levels are expressed on the ratio
scale. A meaningful zero point allows us to state, for example, that profits for firm A are
double those of firm B. Variables such as weight, time, and distance are also measured on
a ratio scale because zero is meaningful.
Unlike nominal- and ordinal-scaled variables (categorical variables), arithmetic oper-
ations are valid on interval- and ratio-scaled variables (numerical variables). In later
chapters, we will calculate summary measures, such as the mean, the median, and
the variance, for numerical variables; we cannot calculate these measures for categorical
variables.
MEASUREMENT SCALES
The observations for any variable can be classified into one of four major measure-
ment scales: nominal, ordinal, interval, or ratio.
∙ Nominal: Observations differ merely by name or label.
∙ Ordinal: Observations can be categorized and ranked; however, differences
between the ranked observations are meaningless.
∙ Interval: Observations can be categorized and ranked, and differences between
observations are meaningful. The main drawback of the interval scale is that
the value of zero is arbitrarily chosen.
∙ Ratio: Observations have all the characteristics of an interval-scaled variable
as well as a true zero point; thus, meaningful ratios can be calculated.
Nominal and ordinal scales are used for categorical variables, whereas interval and
ratio scales are used for numerical variables.
ISTUDY
EXAMPLE 1.2
The owner of a ski resort two hours outside Boston, Massachusetts, is interested
Tween_Survey in serving the needs of the “tween” population (children aged 8 to 12 years old).
He believes that tween spending power has grown over the past few years, and he
wants their skiing experience to be memorable so that they want to return. At the
end of last year’s ski season, he asked 20 tweens the following four questions.
∙ Q1. On your car drive to the resort, which music streaming service was
playing?
∙ Q2. On a scale of 1 to 4, rate the quality of the food at the resort (where 1 is
poor, 2 is fair, 3 is good, and 4 is excellent).
∙ Q3. Presently, the main dining area closes at 3:00 pm. What time do you think
it should close?
∙ Q4. How much of your own money did you spend at the lodge today?
A portion of their responses is shown in Table 1.6. Identify the scale of measure-
ment for each variable used in the survey. Given the tween responses, provide sug-
gestions to the owner for improvement.
TABLE 1.6 Tween Responses to Resort Survey

Tween Music Streaming Food Quality Closing Time Own Money Spent ($)
1 Apple Music 4 5:00 pm 20
2 Pandora 2 5:00 pm 10
⋮ ⋮ ⋮ ⋮ ⋮
20 Spotify 2 4:30 pm 10
SOLUTION:
∙ Q1. Responses for music streaming service are nominal because the observa-
tions differ merely in label. Twelve of the 20 tweens, or 60%, listened to Spo-
tify. If the resort wishes to contact tweens using this means, then it may want
to direct its advertising dollars to this streaming service.
∙ Q2. Food quality responses are on an ordinal scale because we can both
categorize and rank the observations. Eleven of the 20 tweens, or 55%, felt
that the food quality was, at best, fair. Perhaps a more extensive survey that
focuses solely on food quality would reveal the reason for their apparent
dissatisfaction.
∙ Q3. Closing time responses are on an interval scale. We can say that 3:30 pm
is 30 minutes later than 3:00 pm, and 6:00 pm is 30 minutes later than 5:30 pm;
that is, differences between observations are meaningful. The closing time
responses, however, have no apparent zero point. We could arbitrarily define
the zero point at 12:00 am, but ratios are still meaningless. In other words,
it makes no sense to form the ratio 6:00 pm/3:00 pm and conclude that
6:00 pm is twice as long a time period as 3:00 pm. A review of the closing
time responses shows that the vast majority (19 out of 20) would like the
dining area to remain open later.
∙ Q4. The tweens’ responses with respect to their own money spent at the resort
are on a ratio scale. We can categorize and rank observations as well as calculate
meaningful differences. Moreover, because there is a natural zero point, valid
ratios can also be calculated. Seventeen of the 20 tweens spent their own money at
the lodge. It does appear that the discretionary spending of this age group is sig-
nificant. The owner would be wise to cater to some of their preferences.
ISTUDY
Another random document with
no related content on Scribd:
lahonneen kannon kumoon tahi ottaa hengiltä mahdollisesti eteen
sattuvan kirjavan kyyn.
On ollut kevät riistalle suotuisa, tuumii takalistoa kohti verkalleen

astuva vanhus, talvi verrattain vähäluminen ja kesä lämmin sekä
tarpeeksi sateinen. Merkeistä päättäen pitäisi kynä metsissä
löytymän, kynsi kaiken karvallinen Tapiolassa liikkuman.
Tarkasti katselee vorsmestari ympärilleen ja huomioita tekee hän

monenlaisia, milloin mitäkin muistikirjaansa merkiten. Pysähtyy hän
talvisen lumen tekemiä vahinkoja tarkastelemaan — paljonko puita
soreita on kinosten alle sortunut, montako lumen painamana
taittunut. Ja kaatuneen puun alle jääneen mäntynuorukaisen tuon
tuostakin pystyyn auttaa, nuoren näreen kohti valoa kasvamaan
kohottaa. Läheisessä aidassa tarkka silmänsä heti raon keksii, mistä
halkihuulisen jänönpojan on tapana viljelysmaille pujahtaa —
tarttunut on harmaja karva terävään puikkoon raon reunassa.
Huvitettuna hän tuonnempana katselee mäntypistiäisen toukkia,
jotka nuorien petäjien oksilla havuja syövät. Kuin sotamiehet ne
häntä yht'aikaa tervehtivät eturuumiinsa pystyyn nostaen. Sadottain
hän näitä vahinkoeläimiä maahan pudottaa muurahaisten saaliiksi.
Täältä hän lähtee toiskesäisen kulovalkean polttamille maille

katselemaan, missä määrin luonto on jaksanut haavojaan
parannella.
Suunnattoman kasken on eteenpäin rynkäisevä kulovalkea

ryskyen polttanut, laajan kadun tahi mahdottoman leveän mittarin
linjan on humiseva tuli metsää kulkiessaan itselleen aukonut. Aina
tuonne kauvas kohti sinimetsän pehmeäkaarteisia kukkuloita näyttää
nokista kuloa ulottuvan.
Juurellaan komottavia korventuneita mäntyjä, jotka kuivuneina
oksiaan avuttomina ojentelevat, mustia kantoja kuin maasta
nousseita peikkoja kaikkialla, jäkälistä sekä sammalista paljaita
kallioita, kaljuja kiviä tuhansittain ja taas tuhansittain mustassa
maassa — — —. Siellä täällä kaatuneen puun hiiltynyt runko,
oksistoineen kuolleen jättiläis-eläimen luurankoa muistuttaen —
paljon muuta ei tuli näy jälkeensä jättäneen. Kaukaa nähden näyttää
kaikki kuolleelta, kaikkea elämää vailla olevan.
Mutta ei niin lähellä!
Kun vorsmestari likenee, kuulee hän metsäkirvisen visertävän,

kivitaskun maiskuttavan tahi kulorastaan matalaäänisesti tärräävän.
Vilkkaita maakiitäjiä vilahtelee kimaltaen lämpimässä
auringonvalossa ja mustuneen kannon päässä nauttii iso
petokärpänen suojelevasta yhtäläisyydestä. Koivunkannot vihertävät
kuin Aaronin-sauvat ja pientä koivuntainta, hentoa haavanvesaa on
musta maa täynnä. Horsma punaviirinsä kohottaa ikäänkuin sanoen:
»tullaan sitä sysimäelläkin toimeen, eletään vaikka hiilikankahalla»,
keltaisen piiskun kultainen kukkavaltikka nuokkuu kilvan kastikkaan
ruskean röyhyn kanssa ja vihertävinä tuuman korkuisina
havutupsuina kohoavat hentojen männyntaimien lukuisat
sirkkalehdet palaneesta maasta. On tänne jo muitakin tulokkaita
ehtinyt — yhä laajemmalta kulon laitoja valloittavat. Etuvartiona
kulkee kukkiva kanerva, vakoojia lähettää ajuruoho, jonka
sinipunaiset kukkaläikät hietaisen lehmipolun varrella ryydinhajua
tuoksuvat. Ja pääjoukkona tulevat ne, joilla on haivenelliset
hedelmät. Täällä jäkkärät ja kissankäpälät villaisia mykeröitään
tuudittavat, täällä on kangasvuokko kukkinut, täällä härjänsilmät,
keltanot ja monet muut voittoisasti eteenpäin tunkevat. Mutta
kauvempana metsänlaidasta yhä vielä musta erämaa havuneulasten
harvaan peittämänä kolkkona levenee.
Mutta on luonnolla kulovalkean surullisen näyn jälkeen

viehättävääkin näkemistä tarjolla.
Kankaalle vorsmestari nyt nousee, missä nopealentoiset

nuolihaukat kimakasti kii-kii-kii-kii kiijaavat ja kukkiva kanerva uhkaa
ruohottoman polun kokonaan peittää.
Kaikkialle vaatimaton kasvi kukkiensa ujon sinipunan levittää,

ilonväreihin karun hietamaan kilometrittäin pukee ja hunajan
tuoksulla tulijaa tervehtii. Kesän juhlaa viettää kanervikko, punervia
häitä viettävät kanervankukat. Häitä, joissa kimalaiset ja mehiläiset
punakukkain palvelijoina simajuhlassa iloitsevat. Pitkin punapeitettä
lentäen metsästää kuultavasiipinen sudenkorento. Mäyrän nostaman
hietaläjän kohdalla se siipiään liikutellen ilmaan seisahtaa — hetken
liikkumatonna on — äkkiä eteenpäin syöksähtää — kärpäseen iskee
ja saaliineen katoaa.
Kuta kauvemmin vorsmestari polkuaan astelee, jonka poikki

kaarnanhilseessä rapisevat hongat luovat pitkiä varjojaan, sitä
kiihkeämmin odottaa hän linnun siiven kuuluvaksi, teirenemän
äänteleväksi. Ja kankaanlaidan sekametsässä naarasteiri
pyrähtääkin puolanvarsien peitosta, kutkattaa, kutkattaa ja poikiaan
varoittaa — — —.
— Katsotaanpa minkä kokoiset ovat tämän kesän pojat — sanoo

itsekseen vanhus.
Kiireesti hän katkasee muutaman pienen koivun, laatii niistä kojun

kuusen juurelle, jonka runkoa muurahaiset kävelevät kuin ihmiset
vilkasliikkeistä katua. Vihreässä kamarissaan istuen hän »kuikuttaa»
kuin teirenpoika ja vaikenee. Kuikuttaa vorsmestari hetken perästä
taas ja ratokseen hän katselee, miten haapa heittää hyppelevän
valo- ja hämäräverkon maahan. Noin neljännestunnin perästä tulla
hurahtaa emo piilossa kuikuttavaa miestä kohti, istuu pieneen
leppään ihan lähelle ja kaulaa kurkottaen kutsuu, kutsuu poikiaan
kokoon.
Ei kutkata nyt hätäisesti kirjava emo, vaan rauhallisesti se juttelee

kook — — — kook-kook-kook, eikä kauvankaan, niin alkaa
risukossa ripsahdella ja maasta kuulua pientä kuikutusta, kun kirjava
teirenpoika tulla vikittää, että pää nyökkää. Kohta ilmestyy toinenkin,
jolla jo on muutama mustakin höyhen kirjavain seassa. Mustikan syö
kiireesti ja äitiä hakee.
Jo pyrähtää emo maahan ja hiljalleen se alkaa kuljettaa

perhettään kohti tiheimpää viidakkoa. Hetken perästä on näyttämö
tyhjä — mustikanvarsien sekaan on seurue kadonnut.
Tyytyväisenä näytelmään tulee vorsmestari koivumajastaan ja läpi

nuoren sekametsän, missä puut seisovat tiheässä kuin varvut
luudassa, laskeutuu hän alas notkoon, jonka pohjalla kirkas
metsäpuro salamyhkäisesti lirisee. Ensin astuu hän läpi sanajalka-
alueen, missä kuolleenkourat omituisesti hajahtaen aina kainaloihin
saakka kohoavat, painuu sitten läpi pihlajaa, paatsamaa ja haapaa
kasvavan lehdon, missä korkeat keltanot ja täpläiset kämmekkäät
häntä tutusti tervehtivät. Ilmestyy tumma kuusi mäntyjen sekaan,
nurmi katoaa, maa sammaltuu — jo seisoo vaeltaja hiljaa
kiemurtelevan puron varrella, jonka suvantovesiin kuvastuvat korkeat
puut.
Täällä on metsien muhkea hirvi oleskellut, sillä selvinä sen
herttamaiset sorkkain jäljet sammalessa näkyvät. Kolme näkyy
seivässäärtä olleen, kaikki eri kokoa. Ja tuossa ovat maanneet!
Maahan ovat korkeat kastikkaat, metsäkortteet ja hennot vesat
litistäneet.
Joutuu jo vorsmestari korkeaa metsää kulkemaan, missä palokärki

tillii ja männynrungot ruskeina pilareina taivasta tavottelevat, missä
näätä oravan näppää ja kuuset tummina suippenevat kuin
kirkontornit, missä maakotka kohisevin siivin metsästää ja
jäkäläpeitteiset koivut summattoman paksurunkoisiksi paisuvat. Ei
ole kirves täällä ilveksen tuvan ovella vuosimääriin paukkunut ja
kauvan saakin kävellä, ennenkuin korvessa koskemattomassa jotain
ihmisen tekemää jälkeä huomaa. On kuin meren pohjalla käveleisi
— niin on hiljaista salometsän suuressa pilaristossa, sillä ainoastaan
ajottain herää tuuli metsän kannelta soittelemaan. Luonnon korkeata
veisuuta se ison koivun latvassa soittelee, soittelee hetken ja nukkuu
kohta uudelleen viereisen koivun latvaan.
Lepäämään laskeutuu vorsmestari sammaltuneelle kivelle, jonka

pohjoispuoleisessa kyljessä jäkälät rauhallista elämää viettävät —
huvikseen hän siinä katselee vanhoja tuttujaan. Tuossa tuo
rengasmainen, keskeltä mätänevä, laidoilta laajeneva
harmaanvihreä kehäjäkälä. Tuossa peuranjäkälän eri lajit mikä
mitenkin tuhkaharmajana. Tuossa Stereocaulon-jäkälän harmajat
langat kuin uudenvuoden tinat kivelle valettuina — — — pehmeitä,
hillittyjä värejä, jotka silmää viihdyttävät.
Mutta on lintuakin sisällä salon sinisen, on jalan neljän

juoksevatakin. Tuolloin tällöin visertää peipon poikanen, huutaa
kirjava tikka, surisee kurpan viuhuva siipi tahi kotkottaa kirjava
koppelo. Ammottaa ison haavan kyljessä musta reikä ja juurelle
tultuaan huomaa vorsmestari siinä särmäisiä lastuja — pesinyt on
siinä punahuntuinen palokärki. Mutta kun kepillään haavan kylkeen
lyödä kolahuttaa vorsmestari, silloin reikään ilmestyy peljästyneen
siipioravan pyöreä pää. Mustikkamaita kulkiessaan kuulee hän
monen siiven pyrinää yht'aikaa — kohoaa pyyparvi maasta.
Liikkumattomina harmajat linnut piilopaikoissaan istuvat, ei saata
silmä puitten paljoudesta niitä keksiä.
Silloin ottaa vorsmestari pyypiiskunsa esille, asettuu kuusen

juurelle ja soittaa. Ei soita hän nyt uroslinnun kutsuääntä »piip-tiihii-
titititiii», vaan molemmilla pilleillä hän matkii naaraspyyn ääntä,
puhaltaen pitkän »piii». Ja kohta tulla hyristääkin harmaja lintu
pillimiestä kohti, tekee mutkan syrjään ja istuutuu läheiseen
naavakuuseen, joka hellästi ystävänsä partaiseen syliinsä sulkee.
Myhähdellen vorsmestari metsän miellyttävää kanaa katselee,
tähtäilee siihen sauvallaan — otettavana olisi siinä otus! Jo alkaa
pyy oksallaan tepastella, pirisee hetken pit-pit-pit ja pirahtaa sitten
lentoon.
Jo on ilta käsissä, ja hieman kamalaksi alkaa käydä suuren salon

yksinäisyys aurinkoisen aletessa ja notkojen kylmää uhkuessa —
kotimatkalleen kääntyy yksinäinen vaeltaja.
Kaukaa kuuluu kurkien huutoa, ja korkealla hämärtävässä ilmassa

purjehtii kimakasti vinkuen kaareileva hiirihaukka. Syksyä henkii
vaisusti puhaltava iltatuuli, ja alakuloisesti piipittää hömötiainen tsi-
kää-kää-kää.
Pimenevää polkuaan astellessaan vorsmestari kohtaa

iltakävelyllään olevan viirunaamaisen mäyrän, joka äkillisestä
yhtymyksestä peljästyneenä sontiaismetsästyksensä unhottaa ja
oudosti löyhkäten hämärään pakenee. Kankaalla kehrääjä
vorsmestarin eteen polulle lehahtaa, kettu pimeässä haukahtaa ja
liikkumatonna loistaa loppukesän ensimmäinen kiiltomato.
HYLKEENHUUDOSSA
Puolipäivän aurinko paahtaa yli torkkuvan kalastajakylän, jonka

punamaalissa uinuvat rakennukset ja valkokukkaiset puutarhat
välmehtivät helteessä. Hiljaisuutta keskeyttää siimeksessä
märehtivän lampaan kello, janoissaan ammova lehmä tahi
lukemattomain paarmain ja kukkaiskärpästen unettava surina.
Rannan tervaleppien lomitse haamottaa merenlahden sinivesi,
kuvastaen poutaisen taivaan verkalleen purjehtivat valkohattarat ja
tuolloin tällöin kohahtaa loppuun liehtoneen lounaan väsynyt
maininki.
Avonaiseen akkunaani ilmestyvät silakansuomuilla tahrittu

lippalakki ja merituulten kuivaamat, auringon paahtamat kasvot,
akkunanlaudalle pari jäntevää, karvaista kättä, kumpikin yhtä
päivettynyt ja mustakyntinen. Tämä on Sjöström, hylkeenpyytäjä,
joka tervehdittyään kysäsee:
— Jokohan nyt lähdettäisi? Nyt on sopiva ilma!
— Niin hylkeenhuutoonko? No se on tietty!
Pian oltiin matkalla kohti kalastajakylän yhteistä satamaa lahden

kainalossa, missä harmaat ranta-aitat puoliympyrässä törröttävät
toinen toisensa vieressä, rakennettuina kirkkaanvihervään veteen,
jossa mudut ja rautakalat parveilevat ja vihreät levät hiljaa
nuokkuvat.
— Pohjanlahden pohjoisosissa — niin kertoo Sjöström — harmaa

hylje talvensa viettää. Siellä halli tuossa helmi- tahi maaliskuussa
synnyttää ainoan poikansa, jota äiti aution jään päällä imettää noin
kolme tahi neljä viikkoa. Tämän ajan kuluttua luopi poikanen villaisen
syntymäpukunsa, saa vanhempainsa lyhyen karvan ja nyt se äidin
houkuttelemana kohta perehtyy vesi-elämään. Mutta kun Pohjanlahti
kahleistaan vapautuu, ja jäät pohjoistuulten ja merivirtain
kuljettamina ajautuvat tänne etelään päin, silloin seuraavat meren
mahtavat nisäkkäät mukana. Juuri näin Juhannuksen vaiheilla ne
ovat yleiset Ahvenanmeressä. Ja huvittavaa työtä tuo
hylkeenampuminen onkin, kun niitä runsaasti kareille kohoaa…
Mäeltä näemme avaran meren, joka ikäänkuin lämmöstä

väsyneenä hopeanhohtoisena venytteleikse, ja lahden pohjukassa
lukuisat kala-aitat, jotka kaukaa katsoen oikein kylältä näyttävät.
Olkikattojensa alla ne säilyttävät suurimman osan saaristolaisen
irtaimistosta — kaiken minkä merellä tarvitsee.
— Tässä on aittani — ja avaimella, kynnyksen alle kätketyllä,

aukasee Sjöström haulipyssyjen pilkuttaman ja luotien lävistämän
oven. Mutta ulkoseinillä sekä molemmin puolin matalaa
sisäänkäytävää nähdään suuret mustahkot rasvatahrat — sinne
pingoitetaan kuivattavat hylkeennahkat.
Astumme sisään. Puolihämärässä näkyy kilua ja kalua jos

jonkinlaista, outoa ja uutta mannermaalaiselle — kaikki mitä
miellyttävimmässä epäjärjestyksessä.
Leiviskänpainoisia, isoreikäisiä hyljepyssyjä, hyljenaaroja ja
hyljekeihäitä punonaisine jouhiköysineen, keksejä, jääsauvoja.
Orsilla hienorihmaisia jata- ja koukkuverkkoja, silmissä siellä täällä
kuivunut rakkolevätukko. Kiveksiä, lavuksia, verkontupureita
kummallisine puumerkkeineen. Silakkatynnyreitä, astialaudaksia,
vanteita, jalaksellisia kalavasuja. Läjissä kuusenkuorta verkkojen
värjäystä varten ja mukulakiviä varakiveksiksi. Purjeita, mastoja,
airoja, peräsimiä. Pulska rivi haahkan-, allin-, pilkkasiiven-, telkän- ja
koskelonkuvia. Merellä tarvittavia turkkeja, takkeja ja peitteitä. Ja
kaikkialla silakanhajua, kiiltäviä silakansuomuja joka esineessä,
kalpeita silakanpäitä permannolla. Mutta pimeimmässä sopukassa
piilee pari vanhaa kompassia, kuparikattiloita sekä muuta »meren
lahjoittamaa» tavaraa.
Veneeseen sovitetaan matkaan kuuluvat kalut: kompassi,

kaukoputki, kolme hyljepyssyä tarpeineen ja hyljenaara, vesilekkeri
ja eväät, peitteitä ja takkeja.
Airot, purjeet ja peräsin asetetaan paikoilleen, ovi lukitaan,

Sjöström sekä poikansa Konrad istuvat teljoille ja kahden airoparin
tahdikkaasti jysähtäessä sekä liitäväsiipisten tiirain kirkuessa ja
haarapääskysten visertäessä lähtee vene tasaisesti liikkeelle.
Lahden suulla oleva metsäinen saari sivuutetaan ja eteeni

aukenee tyynen meren avara taso, länteenpäin ainoastaan
taivaanrannan rajoittamana.
Ei pelota nyt suuri vesi. Ikäänkuin laiskana kelluen se vain pitkinä,

pehmeinä maininkeina hymyilee — — — ja tunti toisensa perästä
kuluu tasaisen soudun jatkuessa. Taaksemme jäänyt keltainen
hietaranta sekä poukamaa ympäröivä kuusikko painuvat vähitellen
mereen, mutta edessämme alkavat äärimmäiset kalliot kohota
punertavina ja ikäänkuin lämmöstä pehmenneinä.
Aurinko laski, ja länsitaivas hehkui kaikissa iltaruskon

vaihtelevissa väreissä. Kuin kultapohjalle kuvattuna kohosi kaihottu
määräpaikkamme, hallikarimme, vihdoinkin jo lähellä.
Vene kätketään kallion halkeamaan. Varovasti ryömien nousemme

luodon harjalle, missä Sjöström kaukoputkellaan tähystäen
äänettömänä laatii jahtisuunnitelmansa, minun sill'aikaa tehdessä
kasviopillisia huomioita läheisyydessäni.
Pienen lätäkön ympärillä kohtaan miltei koko saariston

kasvillisuuden muutaman neliödekametrin alalle kokoontuneena.
Niittyvillat, variksenmarjat, muuraimet, saraheinät — pohjolan
vaatimattomat lapset Itämeren kasvillisuuden, muhkeiden
kurjenpolvisten, kämmekkäiden, kiviyrttien rinnalla. Pienimmän
liikkeen tehdessä kohoaa ynisevä sääskipilvi edestäni — nekin
Pohjoissuomen muistoja herättäen. Siellä täällä näkyy tyhjä,
variksen tahi lokin tuhoama haahkanpesä — — —.
Vetäydymme alas veneelle taas. Syötyämme vaihdamme uudet

nallit pyssyihin ja sitten saaren lounaiskulmaa kohti konttaamaan,
mutta Konrad jää veneeseen, ollakseen valmiina tarpeen vaatiessa.
Kallion lämpimiin koloihin, suojaan suurten vierinkivien

laittausimme, Sjöström ja minä, niin mukaviin asentoihin kuin
mahdollista. Yksi hyljepyssy, oikea kanuuna mielestäni, asetettiin
suu suoraan eteenpäin, toinen vasemmalle, kolmas oikealle
suunnattiin — halli on näet niin tarkkakuuloinen, ettet saata tuskin
lainkaan siirtää raskasta pyssyä. Heikoimmankin äänen kuultuaan se
silmänräpäyksessä katoaa.
Kaikesta näki, että tässä oli hylkeiden lepopaikka, että hallit olivat
täällä päivää paistattaneet, kuppuroineet, piehtaroineet — karvoja ja
muita jäännöksiä kaikkialla. Olipa vielä »muutakin», kuten tuntuvalla
tavalla kohta tulin huomaamaan — — —.
— No jokohan tuota alotetaan — kuiskaa Sjöström. Ja rintansa

ilmaa täyteen vetäen, jotta se pullistuu kuin palje, hän samassa
päästää pitkän mylvinnän, huutaa kuin sumutorvi, suhistaa
porahuttaa lopuksi ja muuttuu mykäksi, sillä nyt on kaikki puhelu
kielletty.
Ainoastaan nukkuvan meren heikko hengitys kuuluu tuolloin

tällöin, kun unelias maininki juoksee ylös pitkin sileänpunervaa
kalliota, murtuu ja vetäytyy kohisten takaisin. Väljien vesien rajaton
lepo ja sydänkesän hienohämyinen yö, täynnä viihdytystä ja rauhaa,
vaikuttavat kummallisen suloisesti sieluuni. Unhotan mitä varten
venyn tällä lämpöisellä paadella — — —kunnes herään outoon
mölinään. Jo muistan — Sjöströmhän se on, joka halleja
lähenemään houkuttelee.
On peijakkaalla äänivaroja! Hän ulvoo, puhkuu ja porisee milloin

houkuttelevasti mielistellen, milloin valittaen, ikäänkuin pyytäen tahi
anoen, sillä hän tuntee kaikki äänilajit, kaikki vivahdukset hallien
kielessä.
Mutta nyt saan tehdä tuttavuutta tuon »muunkin» kanssa —

suloinen rauhani on mennyt! Nyt ne ovat kimpussani!
Mitkä?
Hyljetäit!
Pitkin käsiä ja kaulaa niitä matelee ja ne purevat, purevat kuin
vimmatut, polttavat kuin tuli! Hilpeän pirullisella mielellä koetan
huvitella itseäni katselemalla niitä suurennuslasini läpi, muistelen
että ne ovat Haematopinus nimisiä — mutta lopputulos on vain
kaiken muun voittava tunne — ne purevat!
Yhä uusia vaivoja!
Jo ovat sääsketkin löytäneet meidät ja nuo pienet lentävät

kuppariämmät panevat yhä useamman sarven ihooni. Tuo suloinen
lämpö käy tukalaksi — on kuuma kuin saunassa. Janottaa — eikä
saa edes tupakoidakaan — hylkeellä on tarkka hajuaistin, väittää
Sjöström, ja sentähden jätettiin tupakkavehkeet veneeseen. Mutta
omat leukansa jauhavat tasaisesti — hän pitää poskessa rakasta
mälliään — — —.
Pieni luotokirvinen sirittää heikosti, lähenee meitä ja pysähtyy

hetkeksi saalista sieppaamaan — hyljetäitäkö syönee vai mitä —
mutta sitten katoaa lintu näkyvistä.
Samassa huomaan sileän merenpinnan särkyvän ja hämäräisenä

häilyvän, veden päälle ilmestyy viiksekäs pää, musta ja pyöreä. Se
häviää, ilmestyy uudelleen — jo lähempänä. Lyhyt kiiltävä kaulakin
näkyy ja päätään käännellen katselee veden peto sinne tänne
pyöreillä, muljottavilla pullosilmillään. Näkyypä sillä olevan
hampaissa iso kalakin — lohiko lie vai mikä — jota se eturaajain
välissä pidellen pureksii. Nykäyksellä muuttaa se sitä ylemmäs sen
mukaan, minkä siitä syö — ja yhä se lähenee.
Kohta on halli vasemman pyssyn ampumalinjassa ja

matalammalla vedellä — rengassilmäisenä, liikkumatonna Sjöström
pyssynperässä tuijottaa, mutta sitten vasen silmänsä ummistuu,
savu pöllähtää, jylhästi hyljepyssy jymähtää. Musta pää ja kiiltävä
rinta katoavat, mutta sijalle jää yhä suureneva rasvaläikkä, ilmoittaen
että luoti on sattunut.
Konrad ilmestyy veneineen ja entisessä asennossaan oleva tyhjä

pyssy näyttää suunnan, mistä eläin on haettava. Naarankoukku
kyljessä se kohta nouseekin pohjasta — ainoastaan talvella on halli
niin lihava, että kuolleena vedenpinnalla kelluu — hinataan kallion
halkeamaan ja ankkuroidaan rantaveteen. Se on hyvin iso,
naarmupäinen halli, monen lohen lopettaja, monessa ottelussa ollut
sukahuulinen uros, vihreäviiksinen vanhus. Luoti on musertanut
ensimmäisen kaulanikaman, sentähden henkikin lähti niinkuin lankaa
poikki leikaten.
Tyhjä pyssy ja Sjöströmin poski ladataan kumpikin —

jälkimmäinen uudella mällipanoksella, ja uusi odotus, huutojen
keskeyttämänä alkaa sekä entinen tulokseton taistelu sääskien ja
hyljetäiden kanssa jatkuu — — —.
Vielä kolme kertaa Konrad soutaa esille rauhallisesta rotkostaan,

noutamaan kolmea nuorta hallia. Yksi ammuttiin miltei maanrajasta,
noin parin kolmen sylen päästä.
Mutta kun ikävöity aurinko vihdoin nousee, on Sjöströmin ääni

lopussa, niin käheä ettei tahdo tulla sanaakaan kurkustaan. Ja minä
niin pehmeänä, että lankean kasvoilleni ja heti nukun veneen
kokkaan päästyäni vihdoin verenhimoisista kiusaajistani.
*****
Aamutuulen herätessä vyörytetään yhteisvoimin nuoret hallit

veneeseen, purjeet nostetaan ja hiljalleen hinataan vanha halli kohti
kotilahden rantaa.
METSOLASSA
Kuulin kerran vanhan jäniksenampujan väittävän, ettei metson

ampuminen haukkuvan piskin avulla ole mitään jaloa huvia. Mutta
mielestäni tällainen metsästys ei ole niinkään ala-arvoiseksi
katsottava, eikä kaikkea urheilua vailla.
Päinvastoin!
Varmaankin usea luonnonystävä myöntänee, että sitä

harjoittaessa helpoimmin tutustuu maamme suureen, jylhään
luontoon.
Tuota täydellistä syventymistä synkän salon helmaan, missä

kankaiden kaunistus ja metsästäjän mielilintu oleskelee, korpien
yksinäisyys ja rämeen raskasmielisyys tovereina, sen rikasta ja
salaperäistä elämää emme opi tuntemiaan silloin, kun »piippovaa
pitkäkorvaa» koirain kiljuessa viljelysmaiden läheisyydessä
ajelemme.
Ei! Pois notkon nuoreen vesakkoon ja kosteihin noroihin, missä

suopursu soiden satuja kertoo, ulos rämeille, missä hetteet heiluvat,
lähteet läikähtävät ja metsäkana kummallisesti ääntelee, siellä alkaa
todellinen, ihmiskädeltä rauhaan jäänyt luonto, salomaa sellainen
kuin minä sitä rakastan! Sinne mieleni tekevi!
Kun metsänimpyet elokuun iltana usvia yli saloniittyjen kutovat,

kun utu valkeana harsona kuuset melkein latvoja myöten kietoo ja
mäkien männyt selväpiirteisinä kohti ruskottavaa länsitaivasta
kohoavat, silloin toivon hyvää metsästysilmaa aamuksi.
Tuossa hienossa hämärässä, salaperäisessä puolivalossa, joka

vallitsee silloin, kun yö alkaa päiväksi muuttua, lähden liikkeelle. Kun
kaste hopeoittaa niityn heinän sekä metsän lehden ja usvat
syleilevät nevan vaivaisia mäntyjä ensimmäisen aamutuulahduksen
hengähtäessä, silloin olen jo metsän peittoon päässyt, kuusikkoa
kumeata tervehtinyt.
Silloin saatan välistä kuulla, miten metso yöpuultaan maahan

kuhahtaa, niin että risahdus kauvas kuuluu hiljaisessa aamuilmassa.
Milloin on syksy, silloin menee jaloin metsälintumme mielellään
ensiksi kytömaan rukiinoraalle. Milloin on elokuu, kuten nyt, silloin
sen löytää nevanlaitaisesta metsävyöhykkeestä, missä mehukkaat
muuraimet keltaisina loistavat, missä rimmen raate ja rämeen
valkosammal, metsäkorte ja suopursu, vaivaiskoivu ja juolukkapaju
vieretysten viihtyvät.
Lomitse naavanpeittämäin kuusien, läpi juolukan- ja

mustikanvarsien kulkee metsämies pienintäkin risahdusta karttaen,
seurassa pystykorvan hallin, joka keltaisena vilkkuu milloin siellä,
milloin täällä puiden takaa. Lähellä nevaa tervehtii häntä metsäkana,
joka valittaen kaakattaa ja poikiaan peittäytymään kehoittaa. Mutta
ennen pitkää koko seurue, yksi toisensa perästä, siipiinsä turvaa ja
häviää näkymättömiin. Ei koirakaan niistä paljon välitä, se tuntee jo
hyvin nuo valkokirjavat linnut, jotka eivät puuhun istahda.
Korpisuossa tirisee pyy, mutta pirahtaa kumminkin kohta lentoon,
kun koira haukkumaan yrittää.
Haavan lehti leplattaa länsituulen herätessä, nouseva aurinko

metsän latvat punaa, kuuset kultihin rakentaa ja pilvettömän sinisenä
kaareutuu taivas — sen minkä sitä voit nähdä, katsoessasi kohti
korkeutta lomitse latvojen.
Mutta nyt hän huomaa ison linnun jäljet painuneina

vaaleanvihreän valkosammalen kosteanpehmeään pintaan! Siitä on
metso kävellä tepastellut ihan äsken — ja samassa kuuluu haukku,
haukku oikein toden perään! Hiiskahtamatta hän lähenee ääntä kohti
ja nyt hän tarvitsee kaikki alkuihmisen vaistot ja aistimet. Sitten hän
huomaa koiran, joka katsomistaan katsoo isoon kuuseen ja haukkuu,
että metsä raikaa — — —. Hän konttaa nelin, lähenee vihdoin
vatsallaan ja on »ampumamatkalla. Piilopaikastaan hän tutkii kuusia,
tarkastaa oksan toisensa perästä löytämättä mitään, ja minuutit
tuntuvat tunneilta! Jo kuulee hän, miten lintu koiralle nokkaa puree,
josta tietää, että vanha metso on edessään, mutta hyvästi näkyy
oksien suojassa lintu istuvan.
Vaan — kärsivällisyyttä!
Jo keksii silmänsä lähellä olevassa vanhassa haavassa jotain,

joka saattaa sydämensä ilosta sykähtämään. Pitkin paksua oksaa
painautuneena, puoleksi lehvän peittämänä näkee hän vanhan
metson koettavan piiloutua tekeytymällä niin pieneksi kuin
mahdollista. Värisevin käsin hän vaihtaa isompihaulisen panoksen
vasempaan piippuun, tähtää, painaa liipasinta ja Tapion kukko tulla
suhahtaa muksahtaen maahan, pieksää sammalia voimakkailla
ruskeilla siivillään ja silmänsä ummistaen tyyntyy vasta, kun halli
hampaansa selkään iskee. Silloin koirakin hänet rauhaan jättää ja
saa palkintonsa — metsonjalan, joka kohta sen hampaissa ratisee.
Sitten he lepäävät hetken, ja hän syö muutaman kourallisen

mustikoita kielensä kostukkeeksi. Ja kun ruvetaan astumaan taas,
tuntuu laukku raskaalta. Sivu mennessään hän pudottaa pyyn
naavakuusesta — — — kauan turhaan silmiään ponnisteltuaan
löytääkseen harmajan linnun harmajasta puusta, ampuu hän sen
vihdoin oksalta, jota moneen kertaan huolellisesti on tarkastellut.
Nevan poikki kulkevan kanervakaarron etelärinteellä löytää hän
pienestä syvennyksestä monta paria poikittain haljenneita
kellanharmaita, ruskeapilkkuisia munankuoria — todistus että
koppelomuori on saanut hautoa rauhassa ja pojat juoksemaan
piipertämään päässeet — jospa lienevät lähelläkin!
Aamuaurinko alkaa lämmittää läpi takin, kaste häviää maasta, ja

ainoastaan silloin tällöin puhaltaa vilvoitellen länsituuli, joka puissa
puhelee, hongissa huokaelee.
Neljä linnun rypemäpaikkaa irtaantuneine höyhenineen kiinnittää

hänen huomiotansa. Hän tarkastelee höyheniä — poikametsoja! Ja
samassa kuuluu rytinää ja siipien pauketta, kun koko poikue
rymähtää lentoon. Omituisesti kotkottaen emäkoppelo perhettään
kutsuu ja samassa alkaa haukku.
Uudestaan metsästysinnon valtaamana hän sykkivin suonin

lähenee, rungolta rungolle käyden, kontaten hiipii. Yhä kuuluu tuo
viettelevä »konk-onk» ja kauvan saa hän katsella, ennenkuin kirjava
koppelo silmään sattuu ja kerran kiusaukseen johdatettuna hän —
laukasee. Vaan kun lintu vihdoin on ammuttu, ei olekaan mieli
voitonriemua täynnä. Miksi ammuit vanhan emon, miksi tapoit
kirjokoppelon! Säästää olisi pitänyt vanha lintu, uuden poikueen
kasvattajaksi — — —.
Tehden yhä laajempia piirejä hakee hän puihin nousseita

poikalintuja. »Jotain» tipahtaa lähellä ole vasta nuoresta koivusta, ja
katsahtaen ylös huomaa hän mustan metsonpojan juuri
lähtemäisillään parempaa turvapaikkaa hakemaan. Mutta samassa
pyssynsä paukahtaa ja oksalta oksalle pudoten mätkähtää nuori lintu
maahan.
Mutta nyt painaa laukku liian tuntuvasti ja suorinta tietä, aurinko

oppaanaan, suuntaa hän matkansa ihmisten ilmoille — — —.

Business Statistics Communicating With Numbers 4E 4Th Edition Sanjiv Jaggia Full Chapter

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Business Statistics Communicating With Numbers 4E 4Th Edition Sanjiv Jaggia Full Chapter

Uploaded by

Copyright:

Available Formats

Business Statistics: Communicating

with Numbers, 4e 4th Edition Sanjiv

Sanjiv Jaggia Alison Kelly

This book is printed on acid-free paper.

Cover Image: McGraw Hill

ISTUDY jag97563_fm_ise iv 11/07/20 12:06 PM

“[Jaggia and Kelly’s text is] an introductory statistics textbook which is

“This introductory statistics book is relevant and approachable. The

“The authors . . . do an excellent job of introducing the concepts.

“This book and its accompanying online resources is ideal for an

“Excellent coverage. . . It is a great book.”

Continuing Key Features

Unique Coverage of Regression Analysis. Relevant coverage of regression without

Integration of Microsoft Excel® and R. Students are taught to develop an understand-

viii B U S I N E S S S TAT I S T I C S WALKTHROUGH

Descriptive: More emphasis on data preparation and visualization

Predictive: Significant rewrite of regression and forecasting

Technology: More reliance on statistical software and Connect

Storytelling: More relevant discussion

ter’s introductory case is presented when the last

Instructors of distance learners may find these introductory_____ cases

SYNOPSIS OF INTRODUCTORY CASE

Dorothy will use the sample information to:

jag16309_ch03_080-123 81 07/22/20 04:00 PM

jag16309_ch06_200-237 229 09/01/20 08:17 PM

jag16309_ch06_200-237 229 07/12/20 01:25 PM

Inclusion of Important Topics

THE SHARPE RATIO

xii B U S I N E S S S TAT I S T I C S WALKTHROUGH

CHAPTeR 5 Discrete Probability Distributions B u S I n e S S S TAT I S T I C S 181

xiv B U S I N E S S S TAT I S T I C S WALKTHROUGH

Features that Go Beyond the

LO 14.4 Interpret goodness-of-fit measures.

514 B U S I n e S S S TAT I S T I C S Conceptual Review

jag16309_ch14_486-517 514 06/17/20 02:00 PM

Study made personal

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

Affordable solutions, Solutions for

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

Study anytime, anywhere “I really liked this

Everything you need in one place

Calendar: owattaphotos/Getty Images

Learning for everyone

New remote proctoring and browser-locking capabilities, hosted by Proctorio within

What Resources are Available for

xviii B U S I N E S S S TAT I S T I C S WALKTHROUGH

MegaStat for Microsoft Excel

E X E RC I S E S 3.4 Integration of Excel Data Sets. A convenient fea-

Calculate the range.

9.4 HYPOtHeSIS teSt FOR tHe

330 B u S I n e S S S tAt I S t I C S 9.4

throughs provide students with step-by-step guidelines

xxiv B U S I N E S S S TAT I S T I C S ACKNOWLEDGEMENTS

2.5 A Stem-and-Leaf Diagram 68

DATA AND DATA PREPARATION 2 CHAPTER 3

Case Study 193 7.5 Statistical Quality Control 259

STATISTICAL INFERENCE CONCERNING 12.2 Chi-Square Test for Independence 416

xxviii B U S I N E S S S TAT I S T I C S CONTENTS

CHAPTER 14 REGRESSION MODELS FOR

REGRESSION MODELS WITH RETURNS, INDEX NUMBERS,

Conceptual Review 616 20.1 Testing a Population Median 684

xxx B U S I N E S S S TAT I S T I C S CONTENTS

LO 1.1 Explain the various data types.

ISTUDY jag97563_fm_ise iv 11/07/20 12:06 PM