100% found this document useful (3 votes)
52 views

[Ebooks PDF] download (eBook PDF) An Introduction to Statistical Methods & Data Analysis 7th full chapters

Introduction

Uploaded by

nyemanneilin
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (3 votes)
52 views

[Ebooks PDF] download (eBook PDF) An Introduction to Statistical Methods & Data Analysis 7th full chapters

Introduction

Uploaded by

nyemanneilin
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 46

Visit https://ebookluna.

com to download the full version and


explore more ebooks

(eBook PDF) An Introduction to Statistical Methods &


Data Analysis 7th

_____ Click the link below to download _____


https://ebookluna.com/product/ebook-pdf-an-
introduction-to-statistical-methods-data-analysis-7th/

Explore and download more ebooks at ebookluna.com


Here are some recommended products that might interest you.
You can download now and explore!

An Introduction to Categorical Data Analysis 3rd Edition


(eBook PDF)

https://ebookluna.com/product/an-introduction-to-categorical-data-
analysis-3rd-edition-ebook-pdf/

ebookluna.com

(eBook PDF) An Introduction to Categorical Data Analysis


by Alan Agresti

https://ebookluna.com/product/ebook-pdf-an-introduction-to-
categorical-data-analysis-by-alan-agresti/

ebookluna.com

(eBook PDF) Qualitative Data Analysis: An Introduction 2nd


Edition

https://ebookluna.com/product/ebook-pdf-qualitative-data-analysis-an-
introduction-2nd-edition/

ebookluna.com

(eBook PDF) An Introduction to Probability and Statistical


Inference 2nd Edition

https://ebookluna.com/product/ebook-pdf-an-introduction-to-
probability-and-statistical-inference-2nd-edition/

ebookluna.com
(eBook PDF) Introduction to Statistics and Data Analysis
6th Edition

https://ebookluna.com/product/ebook-pdf-introduction-to-statistics-
and-data-analysis-6th-edition/

ebookluna.com

(eBook PDF) An Introduction to Statistical Analysis in


Research, Optimized Edition: With Applications in the
Biological and Life Sciences
https://ebookluna.com/product/ebook-pdf-an-introduction-to-
statistical-analysis-in-research-optimized-edition-with-applications-
in-the-biological-and-life-sciences/
ebookluna.com

(eBook PDF) An Introduction to Statistical Problem Solving


in Geography 3rd Edition

https://ebookluna.com/product/ebook-pdf-an-introduction-to-
statistical-problem-solving-in-geography-3rd-edition/

ebookluna.com

(eBook PDF) Handbook of Statistical Analysis and Data


Mining Applications 2nd Edition

https://ebookluna.com/product/ebook-pdf-handbook-of-statistical-
analysis-and-data-mining-applications-2nd-edition/

ebookluna.com

(eBook PDF) An Introduction to Educational Research:


Connecting Methods to Practice

https://ebookluna.com/product/ebook-pdf-an-introduction-to-
educational-research-connecting-methods-to-practice/

ebookluna.com
vi Contents

3.8 Research Study: Controlling for Student Background


in the Assessment of Teaching   119
3. 9 R Instructions  124
3.10 Summary and Key Formulas   124
3.11 Exercises  125

CHAPTER 4 Probability and Probability Distributions 149


4.1 Introduction and Abstract of Research Study­­   149
4.2 Finding the Probability of an Event   153
4.3 Basic Event Relations and Probability Laws   155
4.4 Conditional Probability and Independence   158
4.5 Bayes’ Formula  161
4.6 Variables: Discrete and Continuous   164
4.7 Probability Distributions for Discrete Random Variables   166
4.8 Two Discrete Random Variables: The Binomial and the Poisson   167
4.9 Probability Distributions for Continuous Random Variables   177
4.10 A Continuous Probability Distribution: The Normal Distribution   180
4.11 Random Sampling  187
4.12 Sampling Distributions  190
4.13 Normal Approximation to the Binomial   200
4.14 Evaluating Whether or Not a Population Distribution Is Normal   203
4.15 Research Study: Inferences About Performance-Enhancing Drugs
Among Athletes  208
4.16 R Instructions  211
4.17 Summary and Key Formulas   212
4.18 Exercises  214

PART 4  nalyzing THE Data, Interpreting the


A
Analyses, and Communicating THE Results 231

CHAPTER 5 Inferences About Population Central Values 232


5.1 Introduction and Abstract of Research Study   232
5.2 Estimation of m  235
5.3 Choosing the Sample Size for Estimating m  240
5.4 A Statistical Test for m  242
5.5 Choosing the Sample Size for Testing m  255
5.6 The Level of Significance of a Statistical Test   257
5.7 Inferences About m for a Normal Population, s Unknown  260
5.8 Inferences About m When the Population Is ­Nonnormal and n Is Small:
Bootstrap Methods  269
5.9 Inferences About the Median   275
5.10 Research Study: Percentage of Calories from Fat   280
5.11 Summary and Key Formulas   283
5.12 Exercises  285

CHAPTER 6 Inferences Comparing Two Population Central


Values 300
6.1 Introduction and Abstract of Research Study   300
6.2 Inferences About m1 2 m2: Independent Samples   303

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
Contents vii

6.3 A Nonparametric Alternative:


The Wilcoxon Rank Sum Test   315
6.4 Inferences About m1 2 m2: Paired Data   325
6.5 A Nonparametric Alternative:
The Wilcoxon Signed-Rank Test   329
6.6 Choosing Sample Sizes for Inferences About m1 2 m2  334
6.7 Research Study: Effects of an Oil Spill on Plant Growth   336
6.8 Summary and Key Formulas   341
6.9 Exercises  344

CHAPTER 7 Inferences About Population Variances 366


7.1 Introduction and Abstract of Research Study   366
7.2 Estimation and Tests for a Population Variance   368
7.3 Estimation and Tests for Comparing Two Population Variances   376
7.4 Tests for Comparing t . 2 Population Variances    382
7.5 Research Study: Evaluation of Methods for Detecting E. coli   385
7.6 Summary and Key Formulas   390
7.7 Exercises  391

CHAPTER 8 Inferences About More Than Two Population Central


Values 400
8.1 Introduction and Abstract of Research Study   400
8.2 A Statistical Test About More Than Two Population Means:
An Analysis of Variance    403
8.3 The Model for Observations in a Completely Randomized Design    412
8.4 Checking on the AOV Conditions    414
8.5 An Alternative Analysis: Transformations of the Data    418
8.6 A Nonparametric Alternative: The Kruskal–Wallis Test   425
8.7 Research Study: Effect of Timing on the Treatment
of Port-Wine Stains with Lasers   428
8.8 Summary and Key Formulas   433
8.9 Exercises  435

CHAPTER 9 Multiple Comparisons 445


9.1 Introduction and Abstract of Research Study   445
9.2 Linear Contrasts    447
9.3 Which Error Rate Is Controlled?    454
9.4 Scheffé’s S Method  456
9.5 Tukey’s W Procedure  458
9.6 Dunnett’s Procedure: Comparison of Treatments to a Control   462
9.7 A Nonparametric Multiple-Comparison Procedure   464
9.8 Research Study: Are Interviewers’ Decisions ­Affected by Different
Handicap Types?  467
9.9 Summary and Key Formulas   474
9.10 Exercises  475

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
viii Contents

CHAPTER 10 Categorical Data 482


10.1 Introduction and Abstract of Research Study   482
10.2 Inferences About a Population Proportion p  483
10.3 Inferences About the Difference Between
Two Population Proportions, p1 2 p2  491
10.4 Inferences About Several Proportions:
Chi-Square Goodness-of-Fit Test    501
10.5 Contingency Tables: Tests for Independence
and Homogeneity  508
10.6 Measuring Strength of Relation    515
10.7 Odds and Odds Ratios    517
10.8 Combining Sets of 2 3 2 Contingency Tables  522
10.9 Research Study: Does Gender Bias Exist in the Selection of Students
for Vocational Education?  525
10.10 Summary and Key Formulas   531
10.11 Exercises  533

CHAPTER 11 Linear Regression and Correlation 555


11.1 Introduction and Abstract of Research Study   555
11.2 Estimating Model Parameters    564
11.3 Inferences About Regression Parameters    574
11.4 Predicting New y-Values Using Regression   577
11.5 Examining Lack of Fit in Linear Regression    581
11.6 Correlation  587
11.7 Research Study: Two Methods for Detecting E. coli  598
11.8 Summary and Key Formulas   602
11.9 Exercises  604

CHAPTER 12 Multiple Regression and the General Linear Model 625


12.1 Introduction and Abstract of Research Study    625
12.2 The General Linear Model   635
12.3 Estimating Multiple Regression Coefficients    636
12.4 Inferences in Multiple Regression   644
12.5 Testing a Subset of Regression Coefficients   652
12.6 Forecasting Using Multiple Regression    656
12.7 Comparing the Slopes of Several Regression Lines   658
12.8 Logistic Regression  662
12.9 Some Multiple Regression Theory (Optional)   669
12.10 Research Study: Evaluation of the Performance of an Electric Drill   676
12.11 Summary and Key Formulas   683
12.12 Exercises  685

CHAPTER 13 Further Regression Topics 711


13.1 Introduction and Abstract of Research Study   711
13.2 Selecting the Variables (Step 1)   712
13.3 Formulating the Model (Step 2)   729
13.4 Checking Model Assumptions (Step 3)   745

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
Contents ix

13.5 Research Study: Construction Costs for Nuclear Power Plants   765
13.6 Summary and Key Formulas   772
13.7 Exercises  773

CHAPTER 14 Analysis of Variance for Completely


Randomized Designs 798
14.1 Introduction and Abstract of Research Study   798
14.2 Completely Randomized Design with a Single Factor   800
14.3 Factorial Treatment Structure  805
14.4 Factorial Treatment Structures with an Unequal Number
of Replications  830
14.5 Estimation of Treatment Differences and Comparisons
of Treatment Means  837
14.6 Determining the Number of Replications    841
14.7 Research Study: Development of a Low-Fat Processed Meat   846
14.8 Summary and Key Formulas   851
14.9 Exercises  852

CHAPTER 15 Analysis of Variance for Blocked Designs 865


15.1 Introduction and Abstract of Research Study   865
15.2 Randomized Complete Block Design   866
15.3 Latin Square Design   878
15.4 Factorial Treatment Structure in a Randomized Complete
Block Design  889
15.5 A Nonparametric Alternative—Friedman’s Test   893
15.6 Research Study: Control of Leatherjackets   897
15.7 Summary and Key Formulas   902
15.8 Exercises  904

CHAPTER 16 The Analysis of Covariance 917


16.1 Introduction and Abstract of Research Study   917
16.2 A Completely Randomized Design with One Covariate   920
16.3 The Extrapolation Problem   931
16.4 Multiple Covariates and More Complicated Designs   934
16.5 Research Study: Evaluation of Cool-Season Grasses for Putting
Greens  936
16.6 Summary  942
16.7 Exercises  942

CHAPTER 17 Analysis of Variance for Some Fixed-, Random-,


and Mixed-Effects Models 952
17.1 Introduction and Abstract of Research Study   952
17.2 A One-Factor Experiment with Random Treatment Effects   955
17.3 Extensions of Random-Effects Models   959
17.4 Mixed-Effects Models  967
17.5 Rules for Obtaining Expected Mean Squares   971

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
x Contents

17.6 Nested Factors  981


17.7 Research Study: Factors Affecting Pressure Drops
Across Expansion Joints   986
17.8 Summary  991
17.9 Exercises  992

CHAPTER 18 Split-Plot, Repeated Measures,


and Crossover Designs 1004
18.1 Introduction and Abstract of Research Study   1004
18.2 Split-Plot Designed Experiments   1008
18.3 Single-Factor Experiments with Repeated Measures   1014
18.4 Two-Factor Experiments with Repeated Measures on
One of the Factors   1018
18.5 Crossover Designs  1025
18.6 Research Study: Effects of an Oil Spill on Plant Growth   1033
18.7 Summary  1035
18.8 Exercises  1035

CHAPTER 19 Analysis of Variance for Some Unbalanced


Designs 1050
19.1 Introduction and Abstract of Research Study   1050
19.2 A Randomized Block Design with One or More
Missing Observations  1052
19.3 A Latin Square Design with Missing Data   1058
19.4 Balanced Incomplete Block (BIB) Designs   1063
19.5 Research Study: Evaluation of the Consistency
of Property Assessors  1070
19.6 Summary and Key Formulas   1074
19.7 Exercises  1075

Appendix: Statistical Tables 1085


Answers to Selected Exercises 1125
References 1151
Index 1157

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
PREFACE
INDEX

Intended Audience
An Introduction to Statistical Methods and Data Analysis, Seventh Edition, provides
a broad overview of statistical methods for advanced undergraduate and graduate
students from a variety of disciplines. This book is intended to prepare students to
solve problems encountered in research projects, to make decisions based on data
in general settings both within and beyond the university setting, and finally to
become critical readers of statistical analyses in research papers and in news reports.
The book presumes that the students have a minimal mathematical background
(high school algebra) and no prior course work in statistics. The first 11 chapters
of the textbook present the material typically covered in an introductory statistics
course. However, this book provides research studies and examples that connect
the statistical concepts to data analysis problems that are often encountered in
undergraduate capstone courses. The remaining chapters of the book cover regres-
sion modeling and design of experiments. We develop and illustrate the statistical
techniques and thought processes needed to design a research study or experiment
and then analyze the data collected using an intuitive and proven four-step approach.
This should be especially helpful to graduate students conducting their MS thesis
and PhD dissertation research.

Major Features of Textbook


Learning from Data
In this text, we approach the study of statistics by considering a four-step process
by which we can learn from data:
1. Defining the Problem
2. Collecting the Data
3. Summarizing the Data
4. Analyzing the Data, Interpreting the Analyses, and Communicating
the ­Results

Case Studies
In order to demonstrate the relevance and critical nature of statistics in solving real-
world problems, we introduce the major topic of each chapter using a case study.
The case studies were selected from many sources to illustrate the broad applica-
bility of statistical methodology. The four-step learning from data process is illus-
trated through the case studies. This approach will hopefully assist in overcoming

 xi

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
xii Preface

the natural initial perception held by many people that statistics is just another
“math course.’’ The introduction of major topics through the use of case studies
provides a focus on the central nature of applied statistics in a wide variety of
research and business-related studies. These case studies will hopefully provide the
reader with an enthusiasm for the broad applicability of statistics and the statistical
thought process that the authors have found and used through their many years
of teaching, consulting, and R & D management. The following research studies
­illustrate the types of studies we have used throughout the text.
●● Exit Polls Versus Election Results: A study of why the exit polls
from 9 of 11 states in the 2004 presidential election predicted John
Kerry as the winner when in fact President Bush won 6 of the 11
states.
●● Evaluation of the Consistency of Property Assessors:   A study to
determine if county property assessors differ systematically in their
determination of property values.
●● Effect of Timing of the Treatment of Port-Wine Stains with Lasers:  
A prospective study that investigated whether treatment at a younger
age would yield better results than treatment at an older age.
●● Controlling for Student Background in the Assessment of Teaching:  
An examination of data used to support possible improvements to
the No Child Left Behind program while maintaining the important
concepts of performance standards and accountability.
Each of the research studies includes a discussion of the whys and hows of the
study. We illustrate the use of the four-step learning from data process with each
case study. A discussion of sample size determination, graphical displays of the
data, and a summary of the necessary ingredients for a complete report of the sta-
tistical findings of the study are provided with many of the case studies.

Examples and Exercises


We have further enhanced the practical nature of statistics by using examples and
exercises from journal articles, newspapers, and the authors’ many consulting
experiences. These will provide the students with further evidence of the practical
usages of statistics in solving problems that are relevant to their everyday lives.
Many new exercises and examples have been included in this edition of the book.
The number and variety of exercises will be a great asset to both the instructor and
students in their study of statistics.

Topics Covered
This book can be used for either a one-semester or a two-semester course. Chapters
1 through 11 would constitute a one-semester course. The topics covered would
­include
Chapter 1—Statistics and the scientific method
Chapter 2—Using surveys and experimental studies to gather data
Chapters 3 & 4—Summarizing data and probability distributions
Chapters 5–7—Analyzing data: inferences about central values and
­variances
Chapters 8 & 9—One-way analysis of variance and multiple
comparisons

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
Preface xiii

Chapter 10—Analyzing data involving proportions


Chapter 11—Linear regression and correlation
The second semester of a two-semester course would then include model building
and inferences in multiple regression analysis, logistic regression, design of experi-
ments, and analysis of variance:
Chapters 11–13—Regression methods and model building: multiple re-
gression and the general linear model, logistic regression, and building
­regression models with diagnostics
Chapters 14–19—Design of experiments and analysis of variance: design
concepts, analysis of variance for standard designs, analysis of covari-
ance, random and mixed effects models, split-plot designs, repeated
measures ­designs, crossover designs, and unbalanced designs

Emphasis on Interpretation, not Computation


In the book are examples and exercises that allow the student to study how to
­calculate the value of statistical estimators and test statistics using the definitional
form of the procedure. After the student becomes comfortable with the aspects of
the data the statistical procedure is reflecting, we then emphasize the use of com-
puter software in making computations in the analysis of larger data sets. We provide
output from three major statistical packages: SAS, Minitab, and SPSS. We find that
this approach provides the student with the experience of computing the value of the
procedure using the definition; hence, the student learns the basics b­ ehind each pro-
cedure. In most situations beyond the statistics course, the student should be using
computer software in making the computations for both e­ xpedience and quality of
calculation. In many exercises and examples, the use of the computer allows for more
time to emphasize the interpretation of the ­results of the computations without hav-
ing to expend enormous amounts of time and effort in the ­actual computations.
In numerous examples and exercises, the importance of the following aspects
of hypothesis testing are demonstrated:
1. The statement of the research hypothesis through the summarization
of the researcher’s goals into a statement about population
parameters.
2. The selection of the most appropriate test statistic, including sample
size computations for many procedures.
3. The necessity of considering both Type I and Type II error
rates (a and b) when discussing the results of a statistical test of
hypotheses.
4. The importance of considering both the statistical significance and
the practical significance of a test result. Thus, we illustrate the
importance of estimating effect sizes and the construction of confi-
dence intervals for population parameters.
5. The statement of the results of the statistical test in nonstatistical
jargon that goes beyond the statement ‘‘reject H0’’ or ‘‘fail to
reject H0.’’

New to the Seventh Edition


●● There are instructions on the use of R code. R is a free software package
that can be downloaded from http:/ /lib.stat.cmu.edu/R/CRAN.

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
xiv Preface

Click your choice of platform (Linux, MacOS X, or Windows) for the


precompiled binary distribution. Note the FAQs link to the left for
additional information. Follow the instructions for installing the base
system software (which is all you will need).
●● New examples illustrate the breadth of applications of statistics to
real-world problems.
●● An alternative to the standard deviation, MAD, is provided as a
measure of dispersion in a population/sample.
●● The use of bootstrapping in obtaining confidence intervals and
p-values is discussed.
●● Instructions are included on how to use R code to obtain percentiles
and probabilities from the following distributions: normal, binomial,
Poisson, chi-squared, F, and t.
●● A nonparametric alternative to the Pearson correlation coefficient:
Spearman’s rank correlation, is provided.
●● The binomial test for small sample tests of proportions is presented.
●● The McNemar test for paired count data has been added.
●● The Akaike information criterion and Bayesian information criterion
for variable selection are discussed.

Additional Features Retained from Previous Editions


●● Many practical applications of statistical methods and data analysis
from agriculture, business, economics, education, engineering, medi-
cine, law, political science, psychology, environmental studies, and
sociology have been included.
●● The seventh edition contains over 1,000 exercises, with nearly 400 of
the exercises new.
●● Computer output from Minitab, SAS, and SPSS is provided in
numerous examples. The use of computers greatly facilitates the use
of more sophisticated graphical illustrations of statistical results.
●● Attention is paid to the underlying assumptions. Graphical
procedures and test procedures are provided to determine if assump-
tions have been violated. Furthermore, in many settings, we provide
alternative procedures when the conditions are not met.
●● The first chapter provides a discussion of “What Is Statistics?” We
provide a discussion of why students should study statistics along with
a discussion of several major studies that illustrate the use of statistics
in the solution of real-life problems.

Ancillaries
l Student Solutions Manual (ISBN-10: 1-305-26948-9;
ISBN-13: 978-1-305-26948-4), containing select worked solutions
for problems in the textbook.
l A Companion Website at www.cengage.com/statistics/ott, containing
downloadable data sets for Excel, Minitab, SAS, SPSS, and others,
plus additional resources for students and faculty.

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
Visit https://ebookluna.com
now to explore a diverse
collection of ebooks available
in formats like PDF, EPUB, and
MOBI, compatible with all
devices. Don’t miss the chance
to enjoy exciting offers and
quickly download high-quality
materials in just a few simple
steps!
Preface xv

Acknowledgments
There are many people who have made valuable, constructive suggestions for
the development of the original manuscript and during the preparation of the
subsequent editions. We are very appreciative of the insightful and constructive
comments from the following reviewers:
Naveen Bansal, Marquette University
Kameryn Denaro, San Diego State University
Mary Gray, American University
Craig Leth-Steensen, Carleton University
Jing Qian, University of Massachusetts
Mark Riggs, Abilene Christian University
Elaine Spiller, Marquette University
We are also appreciate of the preparation assistance received from Molly Taylor
and Jay Campbell; the scheduling of the revisions by Mary Tindle, the Senior
Project Manager at Cenveo Publisher Services, who made sure that the book
was completed in a timely manner. The authors of the solutions manual, Soma
Roy, California Polytechnic State University, and John Draper, The Ohio State
University, provided me with excellent input which resulted in an improved set of
exercises for the seventh edition. The person who assisted me the greatest degree
in the preparation of the seventh edition, was Sherry Goldbecker, the copy editor.
Sherry not only corrected my many grammatical errors but also provided rephras-
ing of many sentences which made for a more straight forward explanation of sta-
tistical concepts. The students, who use this book in their statistics classes, will be
most appreciative of Sherry’s many contributions.

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
PART 1

Introduction

Chapter 1 St atistic s a nd the Sc ientific Method

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
1.1 Introduction

CHAPTER 1 1.2
1.3
Why Study Statistics?
Some Current
Applications of Statistics
1.4 A Note to the Student

Statistics and
1.5 Summary
1.6 Exercises

the Scientific
Method

1.1 Introduction
Statistics is the science of designing studies or experiments, collecting data, and
modeling/analyzing data for the purpose of decision making and scientific discov-
ery when the available information is both limited and variable. That is, statistics is
the science of Learning from Data.
Almost everyone, including social scientists, medical researchers, superin-
tendents of public schools, corporate executives, market researchers, engineers,
government employees, and consumers, deals with data. These data could be in the
form of quarterly sales figures, percent increase in juvenile crime, contamination
levels in water samples, survival rates for patients undergoing medical therapy,
census figures, or information that helps determine which brand of car to purchase.
In this text, we approach the study of statistics by considering the four-step process
in Learning from Data: (1) defining the problem, (2) collecting the data, (3) sum-
marizing the data, and (4) analyzing the data, interpreting the analyses, and com-
municating the results. Through the use of these four steps in Learning from Data,
our study of statistics closely parallels the Scientific Method, which is a set of prin-
ciples and procedures used by successful scientists in their p ­ ursuit of knowledge.
The method involves the formulation of research goals, the design of observational
studies and/or experiments, the collection of data, the modeling/analysis of the
data in the context of research goals, and the testing of hypotheses. The conclusion
of these steps is often the formulation of new research goals for a­ nother study.
These steps are illustrated in the schematic given in Figure 1.1.
This book is divided into sections corresponding to the four-step process in
Learning from Data. The relationship among these steps and the chapters of the
book is shown in Table 1.1. As you can see from this table, much time is spent dis-
cussing how to analyze data using the basic methods presented in Chapters 5–19.

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
1.1 Introduction 3

FIGURE 1.1
Scientific Method
Formulate research goal:
Schematic
research hypotheses, models

Formulate new
Make decisions:
research goals:
written conclusions,
new models,
oral presentations
new hypotheses

Design study: Draw inferences:


sample size, variables, Collect data: graphs, estimation,
experimental units, data management hypotheses testing,
sampling mechanism model assessment

TABLE 1.1
Organization of the text The Four-Step Process Chapters

1 Defining the Problem 1 Statistics and the Scientific Method


2 Collecting the Data 2 Using Surveys and Experimental Studies to Gather Data
3 Summarizing the Data 3 Data Description
4 Probability and Probability Distributions
4 Analyzing the Data, 5 Inferences about Population Central Values
Interpreting the Analyses, 6 Inferences Comparing Two Population Central Values
and Communicating 7 Inferences about Population Variances
the Results 8 Inferences about More Than Two Population Central Values
9 Multiple Comparisons
10 Categorical Data
11 Linear Regression and Correlation
12 Multiple Regression and the General Linear Model
13 Further Regression Topics
14 Analysis of Variance for Completely Randomized Designs
15 Analysis of Variance for Blocked Designs
16 The Analysis of Covariance
17 Analysis of Variance for Some Fixed-, Random-, and
Mixed-Effects Models
18 Split-Plot, Repeated Measures, and Crossover Designs
19 Analysis of Variance for Some Unbalanced Designs

However, you must remember that for each data set requiring analysis, someone
has defined the problem to be examined (Step 1), developed a plan for collecting
data to address the problem (Step 2), and summarized the data and prepared the
data for analysis (Step 3). Then following the analysis of the data, the results of the
analysis must be interpreted and communicated either verbally or in written form
to the intended audience (Step 4).
All four steps are important in Learning from Data; in fact, unless the prob-
lem to be addressed is clearly defined and the data collection carried out properly,
the interpretation of the results of the analyses may convey misleading informa-
tion because the analyses were based on a data set that did not address the problem
or that was incomplete and contained improper information. Throughout the text,

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
4 Chapter 1 Statistics and the Scientific Method

we will try to keep you focused on the bigger picture of Learning from Data
through the four-step process. Most chapters will end with a summary section
that emphasizes how the material of the chapter fits into the study of statistics—
Learning from Data.
To illustrate some of the above concepts, we will consider four situations
in which the four steps in Learning from Data could assist in solving a real-world
problem.

1. Problem: Inspection of ground beef in a large beef-processing facility.


A beef-processing plant produces approximately half a million pack-
ages of ground beef per week. The government inspects packages
for possible improper labeling of the packages with respect to the
percent fat in the meat. The inspectors must open the ground beef
package in order to determine the fat content of the ground beef.
The inspection of every package would be prohibitively costly and
time consuming. An alternative approach is to select 250 packages
for inspection from the daily production of 100,000 packages. The
fraction of packages with improper labeling in the sample of 250
packages would then be used to estimate the fraction of packages
improperly labeled in the complete day’s production. If this fraction
exceeds a set specification, action is then taken against the meat
processor. In later chapters, a procedure will be formulated to deter-
mine how well the sample fraction of improperly labeled packages
approximates the fraction of improperly labeled packages for the
whole day’s output.
2. Problem: Is there a relationship between quitting smoking and
gaining weight? To investigate the claim that people who quit
smoking often ­experience a subsequent weight gain, researchers
selected a random sample of 400 participants who had successfully
participated in programs to quit smoking. The individuals were
weighed at the beginning of the program and again 1 year later.
The average change in weight of the participants was an increase of
5 pounds. The investigators concluded that there was evidence that
the claim was valid. We will develop techniques in later chapters to
assess when changes are truly significant changes and not changes
due to random chance.
3. Problem: What effect does nitrogen fertilizer have on wheat production?
For a study of the effects of nitrogen fertilizer on wheat production,
a total of 15 fields was available to the researcher. She randomly
assigned three fields to each of the five nitrogen rates under inves-
tigation. The same variety of wheat was planted in all 15 fields. The
fields were cultivated in the same manner until harvest, and the
number of pounds of wheat per acre was then recorded for each of
the 15 fields. The experimenter wanted to determine the optimal
level of nitrogen to apply to any wheat field, but, of course, she was
limited to running experiments on a limited number of fields. After
determining the amount of nitrogen that yielded the largest produc-
tion of wheat in the study fields, the ­experimenter then concluded
that similar results would hold for wheat fields possessing charac-
teristics somewhat the same as the study fields. Is the experimenter
justified in reaching this conclusion?

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
1.1 Introduction 5

4. Problem: Determining public opinion toward a question, issue,


product, or candidate. Similar applications of statistics are brought
to mind by the frequent use of the New York Times/CBS News,
Washington Post/ABC News, Wall Street Journal/NBC News, Harris,
Gallup/Newsweek, and CNN/Time polls. How can these pollsters
determine the opinions of more than 195 million Americans who are
of voting age? They certainly do not contact every potential voter in
the United States. Rather, they sample the opinions of a small num-
ber of potential voters, perhaps as few as 1,500, to estimate the reac-
tion of every person of voting age in the country. The amazing result
of this process is that if the selection of the voters is done in an unbi-
ased way and voters are asked unambiguous, nonleading questions,
the fraction of those persons contacted who hold a particular opinion
will closely match the fraction in the total population holding that
opinion at a ­particular time. We will supply convincing supportive
evidence of this assertion in subsequent chapters.

These problems illustrate the four-step process in Learning from Data.


First, there was a problem or question to be addressed. Next, for each prob-
lem a study or experiment was proposed to collect meaningful data to solve the
problem. The government meat inspection agency had to decide both how many
packages to inspect per day and how to select the sample of packages from the
total daily output in order to obtain a valid prediction. The polling groups had to
decide how many voters to sample and how to select these individuals in order
to obtain information that is representative of the population of all voters. Simi-
larly, it was necessary to carefully plan how many participants in the weight-gain
study were needed and how they were to be selected from the list of all such
participants. Furthermore, what variables did the researchers have to measure
on each participant? Was it necessary to know each participant’s age, sex, physi-
cal fitness, and other health-related variables, or was weight the only important
variable? The results of the study may not be relevant to the general population
if many of the participants in the study had a particular health condition. In the
wheat experiment, it was important to measure both the soil characteristics of
the fields and the environmental conditions, such as temperature and rainfall, to
obtain results that could be generalized to fields not included in the study. The
design of a study or experiment is crucial to obtaining results that can be general-
ized beyond the study.
Finally, having collected, summarized, and analyzed the data, it is important
to report the results in unambiguous terms to interested people. For the meat
inspection example, the government inspection agency and the personnel in the
beef-processing plant would need to know the distribution of fat content in the
daily production of ground beef. Based on this distribution, the agency could then
impose fines or take other remedial actions against the production facility. Also,
knowledge of this distribution would enable company production personnel to
make adjustments to the process in order to obtain acceptable fat content in their
ground beef packages. Therefore, the results of the statistical analyses cannot
be presented in ambiguous terms; decisions must be made from a well-defined
knowledge base. The results of the weight-gain study would be of vital interest to
physicians who have patients participating in the smoking-cessation program. If
a significant increase in weight was recorded for those individuals who had quit
smoking, physicians would have to recommend diets so that the former smokers

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
6 Chapter 1 Statistics and the Scientific Method

FIGURE 1.2
Population and sample Set of all measurements:
the population

Set of measurements
selected from the
population:
the sample

would not go from one health problem (smoking) to another (elevated blood
pressure due to being overweight). It is crucial that a careful description of the
participants—that is, age, sex, and other health-related information—be included
in the report. In the wheat study, the experiment would provide farmers with
information that would allow them to economically select the optimum amount of
nitrogen required for their fields. Therefore, the report must contain ­information
concerning the amount of moisture and types of soils present on the study fields.
Otherwise, the conclusions about optimal wheat production may not pertain to
farmers growing wheat under considerably different conditions.
To infer validly that the results of a study are applicable to a larger group
population than just the participants in the study, we must carefully define the population
(see Definition 1.1) to which inferences are sought and design a study in which the
sample sample (see Definition 1.2) has been appropriately selected from the designated
population. We will discuss these issues in Chapter 2.

DEFINITION 1.1 A population is the set of all measurements of interest to the sample collector.
(See Figure 1.2.)

DEFINITION 1.2 A sample is any subset of measurements selected from the population.
(See Figure 1.2.)

1.2 Why Study Statistics?


We can think of many reasons for taking an introductory course in statistics. One
reason is that you need to know how to evaluate published numerical facts. Every
person is exposed to manufacturers’ claims for products; to the results of socio-
logical, consumer, and political polls; and to the published results of scientific
research. Many of these results are inferences based on sampling. Some infer-
ences are valid; others are invalid. Some are based on samples of adequate size;
others are not. Yet all these published results bear the ring of truth. Some peo-
ple (particularly statisticians) say that statistics can be made to support almost

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
1.2 Why Study Statistics? 7

anything. Others say it is easy to lie with statistics. Both statements are true. It
is easy, ­purposely or unwittingly, to distort the truth by using statistics when
presenting the results of sampling to the uninformed. It is thus crucial that you
become an ­informed and critical reader of data-based reports and articles.
A second reason for studying statistics is that your profession or employment
may require you to interpret the results of sampling (surveys or experimentation)
or to employ statistical methods of analysis to make inferences in your work. For
example, practicing physicians receive large amounts of advertising describing
the benefits of new drugs. These advertisements frequently display the numerical
­results of experiments that compare a new drug with an older one. Do such data
­really imply that the new drug is more effective, or is the observed difference in
­results due simply to random variation in the experimental measurements?
Recent trends in the conduct of court trials indicate an increasing use of
probability and statistical inference in evaluating the quality of evidence. The use
of statistics in the social, biological, and physical sciences is essential because all
these sciences make use of observations of natural phenomena, through sample
surveys or experimentation, to develop and test new theories. Statistical methods
are employed in business when sample data are used to forecast sales and profit.
In addition, they are used in engineering and manufacturing to monitor product
quality. The sampling of accounts is a useful tool to assist accountants in conduct-
ing audits. Thus, statistics plays an important role in almost all areas of science,
business, and industry; persons employed in these areas need to know the basic
concepts, strengths, and limitations of statistics.
The article “What Educated Citizens Should Know About Statistics and Prob-
ability,” by J. Utts (2003), contains a number of statistical ideas that need to be
understood by users of statistical methodology in order to avoid confusion in the
use of their research findings. Misunderstandings of statistical results can lead to
major errors by government policymakers, medical workers, and consumers of this
information. The article selected a number of topics for discussion. We will sum-
marize some of the findings in the article. A complete discussion of all these topics
will be given throughout the book.
1. One of the most frequent misinterpretations of statistical findings
is when a statistically significant relationship is established between
two variables and it is then concluded that a change in the explana-
tory ­variable causes a change in the response variable. As will be
discussed in the book, this conclusion can be reached only under
very restrictive constraints on the experimental setting. Utts exam-
ined a recent Newsweek article discussing the relationship between
the strength of religious beliefs and physical healing. Utts’ article
discussed the ­problems in reaching the conclusion that the stronger
a patient’s religious beliefs, the more likely the patient would be
cured of his or her ailment. Utts showed that there are ­numerous
other factors involved in a patient’s health and the conclusion that
religious beliefs cause a cure cannot be validly reached.
2. A common confusion in many studies is the difference between
(statistically) significant findings in a study and (practically) signifi-
cant findings. This problem often occurs when large data sets are
involved in a study or experiment. This type of problem will be dis-
cussed in detail throughout the book. We will use a number of exam-
ples that will illustrate how this type of confusion can be avoided by
researchers when reporting the findings of their experimental results.

Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.
Other documents randomly have
different content
led him to forsake. Having returned to her as a babe, and lain at
suck at the brown breasts of the sweet savage, she caresses him
and croons her strange songs in his ears; and to give him pleasure
and sustenance—because the Warrior whom she has long mourned
as lost has ever been the favourite among all her children—she
whispers to him as he lies again in her bosom that he is indeed
divine.
Girt by this forgiveness, and clad in the valour of his divinity once
again, a new and ampler strength is given to the Warrior’s right
hand. He learns for the first time in what manner to use his Sword.
Yet no sooner does he learn so to do than he understands how
Earth, his mother, has deceived him, in order that she may give
pleasure to the favourite among all her children, who has returned to
lie at her brown breasts. And the Warrior blesses her for the
deception, as otherwise he could never have received the strength in
his right hand; and without that strength he could not have learned to
use the sword of his reason. The poem closes in the exaltation of an
infinite pity and tenderness, for the Warrior, blind and spent and
weak, with the Sword all broken and jagged upon his knees, has had
his inquiries answered, and ceasing to struggle and to resent, in his
concern for all his brothers who have not yet learned to use their
swords, he defends their amazing formulas and shibboleths and
extraordinary self-deceptions, whereby they seek to acquire the
strength so to do. And in a passage of tender irony that has no
parallel he entreats his poor brothers to continue to deceive
themselves; and the Warrior concludes with an invocation to Earth,
his mother, for her loving and wise deception, whereby the proudest
of her sons has come to lie at peace.
As the white-haired man continued to read the appreciation of the
poem, which he himself had wrought, thereby re-enacting the
deception of the Mother of the Warrior-Soul, he revealed to the dying
poet how his work in all the wonderful assemblance of its qualities
surpassed any other that had been given to the world. He appraised
the metre which only the highest inspiration would have dared to
employ; he appraised the miraculous blend of gravity, sonority,
sweetness, purity; the ever-gathering range and power of those
mighty cadences, which swept the whole gamut of the emotions as
though they were the strings of a lyre. He showed how the people of
unborn ages would be able to derive stimulus and sanction for their
labours; how the poet’s divine simplicity was such that he who ran
might read; how the official “souls” who infest the groves of Academe
would be able to cherish it for its “art”; how the humblest street-
persons who walked the streets of the great city would be able to
cherish it for its truth. This epic of Man the Warrior ever trampling the
brakes of the Eternal Forest, cutting out the path with the Sword in
his right hand in the fruitless search for that which will reconcile him
to his partial vision, with the nobly pitiful irony of its conclusion that in
the present stage of the Warrior’s development the reconciliation
must be sought by compromise—this epic had in its austere mingling
together of those elements of tragedy which purge man’s nature with
the healing and co-ordinating properties which reconcile diverse and
conflicting factors of experience with the primal belief of Man
Himself, a universal power which had been given to no other poet in
the modern or the ancient world. And the old man concluded with the
prophecy that when “Civilization” itself had sunk to a mere shibboleth
of the remote age of “Reason,” the half-divine, half-barbarous music
of the unknown poet of the Reconciliation, would prove the only via
media between the epoch of ampler vision and that fantastic
shadowography of the long ago when Man seemed other than He
was.
Throughout the reading of this appraisement of his labours, the
blind-eyed poet seemed to vibrate with every word that came upon
his ears. As each phrase uttered by that thin, high, quavering voice
addressed the entranced being of the poet, the frail and broken form
seemed to sway in unison therewith; and the secret and beautiful
smile lurking within the hollows of the cheeks seemed to illuminate
even the sightless eyes, so that poor Dodson, who sat listening
faintly to the old man’s words, was tortured continually by the illusion
that the sight had returned to the eyes of his dying friend.
As the old man, never failing to give expression to his own personal
vindication of that which the poet had wrought, won nearer and
nearer towards the end, the dying man was heard to murmur,
“Courage, Achilles! Courage, Achilles!” for he seemed almost to fear
that consciousness would forsake him before he could realize his
own apotheosis to the full.
When the old man in a kind of triumph and defiance had come to the
end of his task, the radiance upon the poet’s face was starlike in its
lustre.
“Oh, oh, he can see! he can see!” muttered Dodson, in a wild
consternation. “He has the sight in his eyes.”
The poet had stretched forth his weak left hand as though in quest of
something.
He shaped a phrase with his lips, which Dodson had not the power
to understand.
“What does he say?” cried Dodson wildly.
However, the white-haired man appeared to understand. He took
from the table not the carefully written pages from which he had
been reading, but the threepenny reporter’s notebook in which
Dodson’s hastily pencilled criticism had been scribbled. To Dodson’s
profound wonder the old man carried this over to where the poet lay
and placed it in the outstretched left hand. But the hand had not the
power to hold it now.
The poet was heard to mutter some inaudible words.
The old man bore the somewhat unclean threepenny reporter’s
notebook, with its dilapidated green cover, to the lips of the poet,
who pressed them upon it with a half-joyful gesture. In the act he had
ceased to breathe.
It was left to poor Dodson to discover that the act of the divine
clemency had, after many days, been extended to the Warrior-Soul.
The old man was still holding the threepenny reporter’s notebook to
the lips of the mighty dead, when Dodson tore it from his hand.
Clutching it convulsively the young man ran forth of the room and
headlong through the shop. Bare-headed, wild-eyed he reached the
frost-bound, fog-engirdled darkness of the January streets. As he ran
up one street and down another, not knowing nor desiring to know
whither he was bound, yet with that in his clutch ever pressed to his
own white lips, he cried out, “Oh Luney, Luney, I wish now I had
never known you!”
LIX
In the course of the afternoon of the following day, the old man, as
he was clearing away a quantity of débris that choked the fire-grate
in the little room, heard a tap upon the closed shutters of the shop.
Supposing it to proceed from the hand of that blind agent of
providence to whom the world was so much indebted, he left his
occupation and went forth to open the door.
Upon the threshold of the shop he discovered an elderly, grizzled,
grey-bearded man, a total stranger to him. The face of the stranger
was of great resolution.
No sooner had the old man opened the door of the shop and beheld
this unexpected appearance, than the man upon the threshold
looked into his eyes. Suddenly he swept the hat from his head, and
his grey hairs fluttered in the icy January wind.
“I think, sir,” he said in a harsh, strange accent, which yet was that of
awe, “I think, sir, I stand in the presence of the poet.”
The old man recoiled a step from his visitor in mute surprise.
“Forgive me, sir,” said his visitor, “forgive the importunity of the
vulgar, but I am hardly to blame. I have come all the way from
Aberdeen to look upon the poet. You see, I have been a reviewer of
books for the Caledonian Journal for fifty years, but a month ago I
received a book from which my pen has refrained. But I have not
been able to refrain my eyes from its author. To-day, upon my arrival
from Aberdeen, I went direct to the publishers, who at first even
denied an acquaintance with the poet’s name, but ultimately I found
a young man in their office who sent me here.”
“The poet is not I,” said the old man humbly.
The visitor appeared surprised and incredulous.
“If you are not the poet, sir,” he said, “I am sure you are a near
kinsman.”
The old man peered at the grim features of his visitor with his half-
blind eyes. “You appear to be simple and gentle,” he said softly.
“Perhaps you will follow.”
The old man led his visitor into the shop, into the little room, which
was now deserted, and thence up the stairs, into the small chamber
lighted with dim candles, in which the poet lay.
As soon as the visitor beheld that which was therein contained, he
sank to his knees by its side. He remained in that attitude a long
while.
When he arose the aged man was gazing upon him with his half-
blind eyes. They confronted one another like a pair of children.
Suddenly the visitor leaned across the bed in an act of further
homage to the lifeless clay.
“Why do you do that?” said the white-haired man at his side.
“Why do I do this?” said the other, and his powerful spreading
northern speech appeared to strike the walls of the tiny chamber.
“Why do I do this? I am afraid, sir, it must be left to my great great
grandchildren to answer your question.”
THE END

Richard Clay & Sons, Limited, London and Bungay.


NEW NOVELS, 6s.

The Younger Set.


By Robert W. Chambers. Author of “The Fighting
Chance,” “The Reckoning,” “Cardigan,” etc., etc. With 8
Illustrations by G. C. Wilmshurst.
The Measure of the Rule.
By Robert Barr. Author of “The Mutable Many,” “In the
Midst of Alarms,” etc.
In the Shade of the Cloister.
By Arnold Wright.
An impressive story dealing with the inner life of a
Franciscan monastery of our own times, which is
powerfully painted. The whole book is full of striking
features.
The Thornton Device.
By the Hon. Mrs. Grosvenor. Author of “The Bands of
Orion.”
Valerie Upton.
By Anne Douglas Sedgwick. Author of “The Shadow of
Life,” etc.
William Jordan Junior.
By J. C. Snaith. Author of “Broke of Covenden,” “Mistress
Dorothy Marvin,” etc.

The Beaten Road.


By Ellen Glasgow. Author of “The Deliverance,” “The
Wheel of Life,” etc.
A story of modern Virginia, full of interest, in which
there occur some dramatic situations described in
the admirable manner with which readers of “The
Deliverance” will be familiar.
The Square Peg.
By W. E. Norris. Author of “Lord Leonard the Luckless,”
etc.
An admirable story of country-house life, full of
human interest, and written in the inimitable style
that has won for Mr. Norris such a multitude of
readers. The characters are all real people whose
doings will engage the reader’s attention until the
last page is reached.
Marcus Hay.
By Stanley Portal Hyatt.
A tale of frontier life in South-East Africa. To those
who would enjoy a first-class tale of adventure,
written by a man who knows this part of Africa at
first-hand, “Marcus Hay,” gentleman and daring
explorer, can be heartily commended. The story
opens quietly, but soon the reader finds that he
cannot put the book down.
The Helpmate.
By May Sinclair. Author of “The Divine Fire,” etc.
A powerful story, tracing the development of
character through passion and misunderstanding.
The plot of the story is most engrossing. Miss
Sinclair’s last book brought her wide popularity.
A Walking Gentleman.
By James Prior. Author of “Forest Folk,” “Hyssop,” etc.
The story concerns itself with the adventures of a
young nobleman who on the eve of his wedding
abandons his class and takes to the road. The book
has a new note in it—it is fresh and altogether
different from the ruck of most modern novels.
Nicolete. A Novel.
By Evelyn Sharp. Author of “The Youngest Girl in the
School.”
There is no writer whose stories of children, young
and old, reveal more wonderful insight or are more
enjoyable both on account of their truth of
observation and their humour. The heroine of the
story, Nicolete, daughter of a clever, impractical
artist, is a charming study.

A New Book by Mrs. Ernest Thompson Seton


A RECORD OF SPORT, TRAVEL AND NATURE STUDY
IN THE “ROCKIES,” THE CEVENNES, ON THE
OTTAWA, AND IN NORWAY

Nimrod’s Wife
By Grace Gallatin Seton. Author of “A Woman Tenderfoot.”
With numerous Illustrations. Large Crown 8vo. 6s.

Delicia, and other Stories never before


collected.
By Marie Corelli. 3rd Impression. Uniform with “The
Treasure of Heaven.”
New Chronicles of Rebecca.
By Kate Douglas Wiggin. Author of “Rebecca of
Sunnybrook Farm,” “Rose o’ the River,” etc. With
Illustrations by F. C. Yohn.
The Fighting Chance.
By Robert W. Chambers. 5th Impression. Author of “The
Reckoning,” “Maids of Paradise,” etc. With 14 Illustrations
by Fred Pegram.
The Good Comrade.
By Una L. Silberrad. Author of “Curayl,” “The Success of
Mark Wyngate,” etc.
Bachelor Betty.
By Winifred James. 2nd Impression.

Doctor Pons.
By Paul Gwynne. Author of “Marta,” “The Pagan at the
Shrine,” etc.
The “Widda-man.”
By T. Kingston Clarke.
The Three Comrades.
By Gustav Frenssen. Author of “Holyland,” and “Jörn
Uhl.”
Reed Anthony, Cowman.
By Andy Adams. Author of “The Log of a Cowboy,” etc.
Conflict.
By Constance Smedley. Author of “For Heart-o’-Gold,”
“An April Princess,” etc.
The Price of Silence. A Story of New
Orleans.
By M. E. M. Davis. With Illustrations by Griswold Tyng.
THE TREASURE OF HEAVEN
A Romance of Riches
By MARIE CORELLI
With Photogravure Portrait of the Author
Claudius Clear says in the British Weekly:—“It seems to
me to be the best and healthiest of all Miss Corelli’s
books. She is carried along for the greater part of the tale
by a current of pure and high feeling, and she reads a
most wholesome lesson to a generation much tempted to
cynicism—the eternal lesson that love is the prize and the
wealth of life.... The story is full of life from beginning to
end ... it will rank high among the author’s works alike in
merit and popularity.”
The Standard says:—“Miss Corelli gives a brisk, indeed,
a passionate tale of loneliness in search of love, of misery
seeking solace, of the quest of a multi-millionaire for
friendship that is disinterested and affection that has no
purchase price. It is distinctly good to find a preacher with
so great a congregation lifting up her voice against the
selfishness of the time, and urging upon us all the divinity
of faith, charity, and loving-kindness.”
The World and His Wife says:—“It is a pleasant and
absorbing romance, brimful of varied incident, and written
with the alternate vigour and quieter charm that are the
secrets of Miss Corelli’s phenomenal popularity.”
The Bookman says:—“I am going to praise it because I
have found it worth my money. It fulfils the first and most
urgent duty of a novel in having a good story to tell and
telling it interestingly.”

FREE OPINIONS
FREELY EXPRESSED ON
CERTAIN PHASES OF MODERN
SOCIAL LIFE AND CONDUCT
By MARIE CORELLI
“Marmaduke” of Truth says:—“Miss Corelli is a very
clever writer, who has an enormous courage and energy,
and great generosity of mind. In her recently published
book, ‘Free Opinions Freely Expressed,’ these qualities
are especially emphasised, and it is due to Miss Corelli to
acknowledge that she exercises an influence for good in a
period when so few writers are exercising any influence
whatever.”
BY
ERNEST THOMPSON SETON

“I give it as my opinion, that as a writer about Animals,


THOMPSON SETON
CAN’T BE BEATEN.”—Punch.
Animal Heroes.
Being the Histories of a Cat, a Dog, a Pigeon, a Lynx, two
Wolves, and a Reindeer. With over 200 Drawings by the
Author. 6s. net.
The Outlook says:—“Mr. Thompson Seton’s
‘Animal Heroes’ will disappoint none of his readers,
whether old or young, who expect from him a vivid
first-hand description of wild animal life, quickened
by a sense of personal interest in the winged or four-
footed characters with which he brings them into
touch. This is a delightful book for all who care for
animals and animal life, wholly irrespective of age.”
Monarch, the Big Bear of Tallac.
With over 200 Drawings by the Author. 5s. net
Sir Henry Seton Carr says in Vanity Fair:—“Mr.
Thompson Seton can chain the attention of his
readers and carry them along with him in
sympathetic interest for his animal heroes. There is
a human quality about the whole story that makes it
quite impressive. The book is charmingly and
characteristically illustrated.”
The Daily Express says:—“A more charming and
pathetic animal story was never written, even by that
sympathetic student of wild life, Thompson Seton.”
Two Little Savages.
Being the adventures of Two Boys who lived as Indians
and what they learned. With over 300 Drawings by the
Author. 6s. net.
The Daily Chronicle says:—“Let every schoolboy
who wants to be a savage, to understand woodcraft,
to be on intimate terms with things that creep and
swim and fly and lope, demand that his parent shall
give him Mr. Seton’s ‘Two Little Savages.’ Mr. Seton
retains the boyish interest in small and wonderful
things of the forest; he sees all manner of quaint and
absorbing manners in the animals few of us
understand; he knows why the mink fears the cat the
first time, and the cat the mink the second; knows,
too, ‘why the beavers are always so dead sore on
musk rats.’ Moreover, he has a pretty touch with the
pencil, and has spattered drawings of uncommon
vividness and humour about his pages.”
By GEORGE GISSING

The Private Papers of Henry Ryecroft.


[9th Impression
Pocket Edition on thin paper. Cloth Gilt, 2s. 6d. net; full
Limp Lambskin, 3s. 6d. net; Crown 8vo, Cloth Gilt, 6s.
“Mr. Gissing has never written anything more
remarkable.... In many ways it is his best work ...
strikes us as a tour de force.”—The Times.
“The sustained excellence of the writing in this
volume will surprise even his admirers. The pages
that describe natural beauties of scene or of season
are the finest that have been written lately.... The
volume is a great treat. It is the revelation of a
deeply-interesting personality, and it is expressed in
the prose of admirable strength and beauty.”—Daily
Chronicle.
The House of Cobwebs.
[2nd Impression
With an introduction by Thomas Seccombe.
“They are beautiful stories, told with consummate
art, and have a flavour rare in present-day fiction....
It (‘The House of Cobwebs’) is really a masterpiece,
which one is glad to find in the English language.”—
Daily Telegraph.
Veranilda. 6s.
[4th Impression
Mr. H. G. Wells says in The Sphere:—“Gissing’s
maturest, latest and most deliberately-conceived
book ... the book that lay nearest his heart during the
last years of his life.”
Mr. W. L. Courtney, in the Daily Telegraph:—“A
work for which he was eminently fitted by his tastes
and predilections; ‘Veranilda’ is an historical
romance such as we rarely see in our modern
times.”
Dr. William Barry, in The Bookman:—“Fine
workmanship.... It belongs emphatically to literature,
and it cannot fail to give pleasure.”

Will Warburton. A Romance of Real Life. 6s.


[2nd Impression
By MARY JOHNSTON

Pocket Edition on thin paper, with frontispiece. Cloth Gilt,


2s. 6d. net each volume; or full Limp Lambskin, 3s. 6d. net
Crown 8vo Edition, uniform Cloth Gilt, 6s. each.
By Order of the Company.
[14th Edition
“‘By Order of the Company’ has more than fulfilled
the promise of ‘The Old Dominion’ ... a tale of
ingenious exciting adventure, at once catching the
attention, and holding it from first to last.”—The
Globe.
The Old Dominion.
[9th Edition
“Since Thackeray wrote ‘The Virginians’ there has
not been produced a more charming picture of life in
Virginia in the old colonial days than is presented in
Mary Johnston’s romance ‘The Old Dominion.’”—
Daily Mail.
Audrey.
[5th Edition
“A worthy successor to the two other brilliant novels
she has already given us. The whole story is a
beautiful and poetic conception, touched with lights
and shadows of a quiet dry humour and restrained
emotional intensity.... A powerful rememberable
piece of work for which one has nothing but
admiration and praise.”—The Bookman.
Sir Mortimer.
[4th Edition
“‘Sir Mortimer’ will add to the debt owed to her by all
who have read her books.... In the conception of the
plot and its development, and in the creation of
attractive characters, Miss Johnston’s ability is of a
very high order indeed.”—Literary World.
By JOHN FOX

The Little Shepherd of Kingdom Come.


[4th Edition
Illustrated by F. C. Yohn. 6s.
“As pleasant an idyll as you will find in Bret Harte.
Frankly we have not found a dull page in the book.
We doubt if a love scene so pretty as that where
Chad makes his declaration to Margaret has been
written for years.... Jack the sheep-dog is
unforgettable.”—Morning Post.
Crittenden. 6s.
[2nd Edition
“‘Crittenden’ will be found well worth reading, both
as a story and as history.... It is written in an easy,
vigorous style, and the interest never flags.”—
Academy.
Blue-Grass and Rhododendron.
Outdoor Life in old Kentucky.
Illustrated. Cloth gilt extra. 6s. net.
“A most fascinating book about Kentucky.”—Daily
Telegraph.
“Vivid pictures of the Blue-Grass people at work, at
play, at religion, at politics.”—Christian World.
Christmas Eve on Lonesome.
With Coloured Illustrations. Cr. 8vo. 3s. 6d. net.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and


personal growth!

ebookluna.com

You might also like