Sampling 3ed. Edition Steven K.

Thompson - PDF
Download (2025)

https://ebookultra.com/download/sampling-3ed-edition-steven-k-
thompson/

Visit ebookultra.com today to download the complete set of

ebooks or textbooks
We have selected some products that you may be interested in
Click the link to download now or visit ebookultra.com
for more options!.

The Psychology of Genocide 1st Edition Steven K. Baum

https://ebookultra.com/download/the-psychology-of-genocide-1st-
edition-steven-k-baum/

Rorschach Assessment of the Personality Disorders 1st

Edition Steven K. Huprich

https://ebookultra.com/download/rorschach-assessment-of-the-
personality-disorders-1st-edition-steven-k-huprich/

Python cookbook 3ed. Edition Beazley D.

https://ebookultra.com/download/python-cookbook-3ed-edition-beazley-d/

Fundamental neuroscience 3ed Edition Squire L.R.

https://ebookultra.com/download/fundamental-neuroscience-3ed-edition-
squire-l-r/
Erp Demystified 3Ed 3rd Edition Leon

https://ebookultra.com/download/erp-demystified-3ed-3rd-edition-leon/

Fundamental food microbiology 3ed Edition Ray B.

https://ebookultra.com/download/fundamental-food-microbiology-3ed-
edition-ray-b/

Settled Asbestos Dust Sampling and Analysis First Edition

Hays

https://ebookultra.com/download/settled-asbestos-dust-sampling-and-
analysis-first-edition-hays/

Introduction to biomedical engineering 3ed. Edition

Enderle J.

https://ebookultra.com/download/introduction-to-biomedical-
engineering-3ed-edition-enderle-j/

Encyclopedia of Chromatography 3ed. Edition Cazes J. (Ed.)

https://ebookultra.com/download/encyclopedia-of-chromatography-3ed-
edition-cazes-j-ed/
Sampling 3ed. Edition Steven K. Thompson Digital
Instant Download
Author(s): Steven K. Thompson
ISBN(s): 9780470402313, 0470402318
Edition: 3ed.
File Details: PDF, 1.60 MB
Year: 2012
Language: english
Sampling
Sampling

Third Edition

STEVEN K. THOMPSON
Simon Fraser University

A JOHN WILEY & SONS, INC., PUBLICATION

Copyright © 2012 by John Wiley & Sons, Inc. All rights reserved.

Published by John Wiley & Sons, Inc., Hoboken, New Jersey.

Published simultaneously in Canada.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form
or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as
permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior
written permission of the Publisher, or authorization through payment of the appropriate per-copy fee
to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400,
fax (978) 750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission
should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street,
Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at
http://www.wiley.com/go/permission.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts
in preparing this book, they make no representations or warranties with respect to the accuracy or
completeness of the contents of this book and specifically disclaim any implied warranties of
merchantability or fitness for a particular purpose. No warranty may be created or extended by sales
representatives or written sales materials. The advice and strategies contained herein may not be
suitable for your situation. You should consult with a professional where appropriate. Neither the
publisher nor author shall be liable for any loss of profit or any other commercial damages, including
but not limited to special, incidental, consequential, or other damages.

For general information on our other products and services or for technical support, please contact our
Customer Care Department within the United States at (800) 762-2974, outside the United States at
(317) 572-3993 or fax (317) 572-4002.

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print
may not be available in electronic formats. For more information about Wiley products, visit our web
site at www.wiley.com.

Library of Congress Cataloging-in-Publication Data:

Thompson, Steven K., 1945–

Sampling / Steven K. Thompson. — 3rd ed.
p.cm. — (Wiley series in probability and statistics ; 755)
Includes index.
ISBN 978-0-470-40231-3 (hardback)
1 Sampling (Statistics) I. Title.
QA276.6.T58 2012
519.5 2—dc23
2011028944

Printed in the United States of America

10 9 8 7 6 5 4 3 2 1
 Contents

Preface xv

Preface to the Second Edition xvii

Preface to the First Edition xix

1 Introduction 1
1.1 Basic Ideas of Sampling and Estimation, 2
1.2 Sampling Units, 4
1.3 Sampling and Nonsampling Errors, 5
1.4 Models in Sampling, 5
1.5 Adaptive and Nonadaptive Designs, 6
1.6 Some Sampling History, 7

PART I BASIC SAMPLING 9

2 Simple Random Sampling 11

2.1 Selecting a Simple Random Sample, 11
2.2 Estimating the Population Mean, 13
2.3 Estimating the Population Total, 16
2.4 Some Underlying Ideas, 17
2.5 Random Sampling with Replacement, 19
2.6 Derivations for Random Sampling, 20
2.7 Model-Based Approach to Sampling, 22
2.8 Computing Notes, 26
Entering Data in R, 26
Sample Estimates, 27

v
vi contents

Simulation, 28
Further Comments on the Use of Simulation, 32
Exercises, 35

3 Confidence Intervals 39
3.1 Confidence Interval for the Population Mean or Total, 39
3.2 Finite-Population Central Limit Theorem, 41
3.3 Sampling Distributions, 43
3.4 Computing Notes, 44
Confidence Interval Computation, 44
Simulations Illustrating the Approximate Normality of a
Sampling Distribution with Small n and N , 45
Daily Precipitation Data, 46
Exercises, 50

4 Sample Size 53
4.1 Sample Size for Estimating a Population Mean, 54
4.2 Sample Size for Estimating a Population Total, 54
4.3 Sample Size for Relative Precision, 55
Exercises, 56

5 Estimating Proportions, Ratios, and Subpopulation Means 57

5.1 Estimating a Population Proportion, 58
5.2 Confidence Interval for a Proportion, 58
5.3 Sample Size for Estimating a Proportion, 59
5.4 Sample Size for Estimating Several Proportions
Simultaneously, 60
5.5 Estimating a Ratio, 62
5.6 Estimating a Mean, Total, or Proportion of a
Subpopulation, 62
Estimating a Subpopulation Mean, 63
Estimating a Proportion for a Subpopulation, 64
Estimating a Subpopulation Total, 64
Exercises, 65

6 Unequal Probability Sampling 67

6.1 Sampling with Replacement: The Hansen–Hurwitz
Estimator, 67
6.2 Any Design: The Horvitz–Thompson Estimator, 69
6.3 Generalized Unequal-Probability Estimator, 72
contents vii

6.4 Small Population Example, 73

6.5 Derivations and Comments, 75
6.6 Computing Notes, 78
Writing an R Function to Simulate a Sampling Strategy, 82
Comparing Sampling Strategies, 84
Exercises, 88

PART II MAKING THE BEST USE OF SURVEY DATA 91

7 Auxiliary Data and Ratio Estimation 93

7.1 Ratio Estimator, 94
7.2 Small Population Illustrating Bias, 97
7.3 Derivations and Approximations for the Ratio Estimator, 99
7.4 Finite-Population Central Limit Theorem for the Ratio
Estimator, 101
7.5 Ratio Estimation with Unequal Probability Designs, 102
7.6 Models in Ratio Estimation, 105
Types of Estimators for a Ratio, 109
7.7 Design Implications of Ratio Models, 109
7.8 Computing Notes, 110
Exercises, 112

8 Regression Estimation 115

8.1 Linear Regression Estimator, 116
8.2 Regression Estimation with Unequal Probability
Designs, 118
8.3 Regression Model, 119
8.4 Multiple Regression Models, 120
8.5 Design Implications of Regression Models, 123
Exercises, 124

9 The Sufficient Statistic in Sampling 125

9.1 The Set of Distinct, Labeled Observations, 125
9.2 Estimation in Random Sampling with Replacement, 126
9.3 Estimation in Probability-Proportional-to-Size Sampling, 127
9.4 Comments on the Improved Estimates, 128

10 Design and Model 131

10.1 Uses of Design and Model in Sampling, 131
viii contents

10.2 Connections between the Design and Model Approaches, 132

10.3 Some Comments, 134
10.4 Likelihood Function in Sampling, 135

PART III SOME USEFUL DESIGNS 139

11 Stratified Sampling 141

11.1 Estimating the Population Total, 142
With Any Stratified Design, 142
With Stratified Random Sampling, 143
11.2 Estimating the Population Mean, 144
With Any Stratified Design, 144
With Stratified Random Sampling, 144
11.3 Confidence Intervals, 145
11.4 The Stratification Principle, 146
11.5 Allocation in Stratified Random Sampling, 146
11.6 Poststratification, 148
11.7 Population Model for a Stratified Population, 149
11.8 Derivations for Stratified Sampling, 149
Optimum Allocation, 149
Poststratification Variance, 150
11.9 Computing Notes, 151
Exercises, 155

12 Cluster and Systematic Sampling 157

12.1 Primary Units Selected by Simple Random Sampling, 159
Unbiased Estimator, 159
Ratio Estimator, 160
12.2 Primary Units Selected with Probabilities Proportional to
Size, 161
Hansen–Hurwitz (PPS) Estimator, 161
Horvitz–Thompson Estimator, 161
12.3 The Basic Principle, 162
12.4 Single Systematic Sample, 162
12.5 Variance and Cost in Cluster and Systematic
Sampling, 163
12.6 Computing Notes, 166
Exercises, 169
contents ix

13 Multistage Designs 171

13.1 Simple Random Sampling at Each Stage, 173
Unbiased Estimator, 173
Ratio Estimator, 175
13.2 Primary Units Selected with Probability Proportional to
Size, 176
13.3 Any Multistage Design with Replacement, 177
13.4 Cost and Sample Sizes, 177
13.5 Derivations for Multistage Designs, 179
Unbiased Estimator, 179
Ratio Estimator, 181
Probability-Proportional-to-Size Sampling, 181
More Than Two Stages, 181
Exercises, 182

14 Double or Two-Phase Sampling 183

14.1 Ratio Estimation with Double Sampling, 184
14.2 Allocation in Double Sampling for Ratio Estimation, 186
14.3 Double Sampling for Stratification, 186
14.4 Derivations for Double Sampling, 188
Approximate Mean and Variance: Ratio Estimation, 188
Optimum Allocation for Ratio Estimation, 189
Expected Value and Variance: Stratification, 189
14.5 Nonsampling Errors and Double Sampling, 190
Nonresponse, Selection Bias, or Volunteer Bias, 191
Double Sampling to Adjust for Nonresponse:
Callbacks, 192
Response Modeling and Nonresponse Adjustments, 193
14.6 Computing Notes, 195
Exercises, 197

PART IV METHODS FOR ELUSIVE AND HARD-TO-DETECT

POPULATIONS 199

15 Network Sampling and Link-Tracing Designs 201

15.1 Estimation of the Population Total or Mean, 202
Multiplicity Estimator, 202
Horvitz–Thompson Estimator, 204
x contents

15.2 Derivations and Comments, 207

15.3 Stratification in Network Sampling, 208
15.4 Other Link-Tracing Designs, 210
15.5 Computing Notes, 212
Exercises, 213

16 Detectability and Sampling 215

16.1 Constant Detectability over a Region, 215
16.2 Estimating Detectability, 217
16.3 Effect of Estimated Detectability, 218
16.4 Detectability with Simple Random Sampling, 219
16.5 Estimated Detectability and Simple Random Sampling, 220
16.6 Sampling with Replacement, 222
16.7 Derivations, 222
16.8 Unequal Probability Sampling of Groups with Unequal
Detection Probabilities, 224
16.9 Derivations, 225
Exercises, 227

17 Line and Point Transects 229

17.1 Density Estimation Methods for Line Transects, 230
17.2 Narrow-Strip Method, 230
17.3 Smooth-by-Eye Method, 233
17.4 Parametric Methods, 234
17.5 Nonparametric Methods, 237
Estimating f (0) by the Kernel Method, 237
Fourier Series Method, 239
17.6 Designs for Selecting Transects, 240
17.7 Random Sample of Transects, 240
Unbiased Estimator, 241
Ratio Estimator, 243
17.8 Systematic Selection of Transects, 244
17.9 Selection with Probability Proportional to Length, 244
17.10 Note on Estimation of Variance for the Kernel
Method, 246
17.11 Some Underlying Ideas about Line Transects, 247
Line Transects and Detectability Functions, 247
Single Transect, 249
Average Detectability, 249
contents xi

Random Transect, 250

Average Detectability and Effective Area, 251
Effect of Estimating Detectability, 252
Probability Density Function of an Observed Distance, 253
17.12 Detectability Imperfect on the Line or Dependent on
Size, 255
17.13 Estimation Using Individual Detectabilities, 255
Estimation of Individual Detectabilities, 256
17.14 Detectability Functions other than Line Transects, 257
17.15 Variable Circular Plots or Point Transects, 259
Exercise, 260

18 Capture–Recapture Sampling 263

18.1 Single Recapture, 264
18.2 Models for Simple Capture–Recapture, 266
18.3 Sampling Design in Capture–Recapture: Ratio Variance
Estimator, 267
Random Sampling with Replacement of Detectability
Units, 269
Random Sampling without Replacement, 270
18.4 Estimating Detectability with Capture–Recapture
Methods, 271
18.5 Multiple Releases, 272
18.6 More Elaborate Models, 273
Exercise, 273

19 Line-Intercept Sampling 275

19.1 Random Sample of Lines: Fixed Direction, 275
19.2 Lines of Random Position and Direction, 280
Exercises, 282

PART V SPATIAL SAMPLING 283

20 Spatial Prediction or Kriging 285

20.1 Spatial Covariance Function, 286
20.2 Linear Prediction (Kriging), 286
20.3 Variogram, 289
20.4 Predicting the Value over a Region, 291
20.5 Derivations and Comments, 292
xii contents

20.6 Computing Notes, 296

Exercise, 299

21 Spatial Designs 301

21.1 Design for Local Prediction, 302
21.2 Design for Prediction of Mean of Region, 302

22 Plot Shapes and Observational Methods 305

22.1 Observations from Plots, 305
22.2 Observations from Detectability Units, 307
22.3 Comparisons of Plot Shapes and Detectability Methods, 308

PART VI ADAPTIVE SAMPLING 313

23 Adaptive Sampling Designs 315

23.1 Adaptive and Conventional Designs and Estimators, 315
23.2 Brief Survey of Adaptive Sampling, 316

24 Adaptive Cluster Sampling 319

24.1 Designs, 321
Initial Simple Random Sample without Replacement, 322
Initial Random Sample with Replacement, 323
24.2 Estimators, 323
Initial Sample Mean, 323
Estimation Using Draw-by-Draw Intersections, 323
Estimation Using Initial Intersection Probabilities, 325
24.3 When Adaptive Cluster Sampling Is Better than Simple
Random Sampling, 327
24.4 Expected Sample Size, Cost, and Yield, 328
24.5 Comparative Efficiencies of Adaptive and Conventional
Sampling, 328
24.6 Further Improvement of Estimators, 330
24.7 Derivations, 333
24.8 Data for Examples and Figures, 336
Exercises, 337

25 Systematic and Strip Adaptive Cluster Sampling 339

25.1 Designs, 341
25.2 Estimators, 343
contents xiii

Initial Sample Mean, 343

Estimator Based on Partial Selection Probabilities, 344
Estimator Based on Partial Inclusion Probabilities, 345
25.3 Calculations for Adaptive Cluster Sampling Strategies, 347
25.4 Comparisons with Conventional Systematic and Cluster
Sampling, 349
25.5 Derivations, 350
25.6 Example Data, 352
Exercises, 352

26 Stratified Adaptive Cluster Sampling 353

26.1 Designs, 353
26.2 Estimators, 356
Estimators Using Expected Numbers of Initial
Intersections, 357
Estimator Using Initial Intersection Probabilities, 359
26.3 Comparisons with Conventional Stratified Sampling, 362
26.4 Further Improvement of Estimators, 364
26.5 Example Data, 367
Exercises, 367

Answers to Selected Exercises 369

References 375

Author Index 395

Subject Index 399

 Preface

One change with this edition of Sampling is that I have included sections of
computing notes for sample selection, calculation of estimates, and simulations.
These computations are illustrated using the statistical programming language R.
In doing this I have avoided the use of specialized packages for specific complex
designs, choosing instead to show simple calculations and sampling procedures
from scratch using a few basic functions. The purpose of these sections is as much
for understanding of sampling ideas as for easy ways to select samples and calcu-
late estimates. Other software than R can, of course, be used for the same purpose.
The advantages of R include: it is a free and open source, is widely supported by
the statistical and other research communities, is available to anyone, and is easily
installed on a computer with any of the common operating systems, including
Windows, Macintosh OS X, Linux, and other types of Unix. The syntax of R tends
to read like generic code and conveys the thinking that goes along with calculations
rather than serving as a magic box. R is interactive and has very nice graphics.
Once one learns how to select a sample with a given type of design and to pro-
duce various types of estimates using the sample data from the design, it is an easy
step to wrap that procedure into a simulation of a sampling strategy. Much of the
attention of the computing sections is devoted to the simulation of sampling strate-
gies. The idea is to construct a “population” in the computer as much as possible
like the real one which needs to be sampled. With this artificial but more-or-less
realistic population, the sampling strategy is then carried out many times. So on
each of the runs a sample is selected using the design, and estimates are calculated
from the sample data obtained. The distribution of these estimates over the many
runs is the sampling distribution. It depends as much on the sampling design and
estimation procedure chosen as upon the characteristics of the population. In this
way one prospective sampling strategy can be evaluated in comparison to others
before committing to one to use in the field. In addition to providing a practical
way to evaluate and improve potential sampling strategies, simulations of this kind
can give an understanding that is right at the heart of sampling.
Some new examples have been added to this edition. New figures have been
added, in particular illustrating the ideas of sampling distributions and the results
xv
xvi preface

of various types of simulations. Numerous incremental improvements and the odd

new section have been added.
I would like to thank especially the students in my classes and colleagues at other
institutions who have helped with corrections of typographical errors and other
improvements. I would like to thank Susanne Steitz-Filler and Stephen Quigley
at John Wiley & Sons for encouragement in preparation of this edition. Research
support for my work in the area of sampling has been provided by the Natural Sci-
ences and Engineering Research Council, the National Center for Health Statistics,
Centers for Disease Control and Prevention, the U.S. Census Bureau, the National
Institutes of Health, and the National Science Foundation.

Steven K. Thompson
Simon Fraser University
British Columbia
 Preface to the Second Edition

The Second Edition retains the general organization of the first, but incorporates
new material interspersed throughout the text. For example, model-based ideas
and alternatives are included from the earliest chapters, including those on simple
random sampling and stratified sampling, rather than suddenly appearing along
with ratio and regression estimation methods as has been traditional. Estimation
methods deriving from a combination of design and model considerations receive
added attention in this edition. Some useful ideas from the ever-developing theory
of sampling are briefly described in the chapters on making the most of survey data.
Among the added sections is an expanded description of methods for adjusting
for nonsampling errors. A wider discussion of link-tracing designs for sampling
hidden human populations—or the Internet—has been added to the chapter on
network sampling. New developments in the rapidly expanding field of adaptive
sampling are briefly summarized.
Additional numerical examples, as well as exercises, have been added. A number
of additional derivations of results have been tucked into the later parts of chapters.
A brief history of sampling has been added to the introduction.
I would like to express my thanks and appreciation to the many people who
have so generously shared with me their views on sampling theory and methods
in discussions, collaborations, and visits to field sites. They include my colleagues
at The Pennsylvania State University and those in the wider research community
of sampling and statistics, as well as researchers in other fields such as ecology,
biology, environmental science, computer science, sociology, anthropology, ethnog-
raphy, and the health sciences. I would like to thank my editor Steve Quigley and
editorial program coordinator Heather Haselkorn at John Wiley & Sons for their
encouragement and assistance with this project. Research support for my work has
been provided by grants from the National Science Foundation (DMS-9626102)
and the National Institutes of Health (R01 DA09872).

Steven K. Thompson
University Park, Pennsylvania

xvii
 Preface to the First Edition

This book covers the basic and standard sampling design and estimation methods
and, in addition, gives special attention to methods for populations that are inher-
ently difficult to sample, elusive, rare, clustered, or hard to detect. It is intended as
a reference for scientific researchers and others who use sampling and as a textbook
for a graduate or upper-level undergraduate course in sampling.
The twenty-six chapters of the book are organized into six parts. Part I cov-
ers basic sampling from simple random sampling to unequal probability sampling.
Part II treats the use of auxiliary data with ratio and regression estimation and
looks at the ideas of sufficient data and of model and design in practical sam-
pling. Part III covers major useful designs including stratified, cluster, systematic,
multistage, double, and network sampling. Part IV examines detectability methods
for elusive populations: Basic problems in detectability, visibility, and catchabil-
ity are discussed and specific methods of line transects, variable circular plots,
capture–recapture, and line-intercept sampling are covered. Part V concerns spatial
sampling, with the prediction or “kriging” methods of geostatistics, considerations
of efficient spatial designs, and comparisons of different observational methods
including plot shapes and detection aspects. Part VI introduces adaptive sam-
pling designs, in which the sampling procedure depends on what is observed
during the survey; for example, sampling effort may be increased in the vicin-
ity of high observed abundance. The adaptive cluster sampling designs described
can be remarkably effective for sampling rare, clustered populations, which by
conventional methods are notoriously difficult to sample.
Researchers faced with such problems as estimating the abundance of an animal
population or an elusive human population, predicting the amount of mineral or
fossil-fuel resource at a new site, or estimating the prevalence of a rare disease
must be aware that the most effective methods go beyond the material traditionally
found in sampling books. At the same time, such researchers may not be aware of
the potential usefulness of some of the relatively recent developments in sampling
theory and methods—such as network sampling, adaptive sampling designs, and
generalized ratio and regression estimation with unequal probability designs. For
xix
xx preface to the first edition

these reasons, the selection of topics covered in this book is wider than has been
traditional for sampling texts.
Some important sampling methodologies have developed largely in particular
fields—such as ecology, geology, or health sciences—seemingly in isolation from
the mainstream of statistical sampling theory. In the chapters on such methods, I
have endeavored to bring out the connections with and the advantages to be gained
from basic sampling design, estimation, and prediction results. Thus, for instance,
in the chapters on detectability methods associated in particular with ecological
sampling, sampling design is emphasized. In the chapter on the prediction or krig-
ing methods associated with geostatistics, the connection to regression estimation
results is noted. In the chapter on network sampling, originally associated with
epidemiological surveys, the notation has been simplified and connections to basic
unequal probability sampling estimators are observed.
Although the range of topics in this book is for the above-noted reasons con-
siderably wider than has been traditional for sampling texts, it has been necessary,
in order to keep the book of the desired size, to be selective in what to include.
To the reader for whom an additional topic would have been particularly helpful,
I can only offer the recompense of the references cited throughout the text to give
access to the wider literature in sampling.
My immediate purposes in writing this book were to provide a text for graduate
and upper-level undergraduate courses in sampling at the University of Alaska
Fairbanks and at the University of Auckland and to provide a manual of useful
sampling and estimation methods for researchers with whom I had worked on
various projects in a variety of scientific fields. No available manual or text covered
the range of topics of interest to these people.
In my experience the backgrounds of the researchers and students interested
in sampling topics have been extremely diverse: While some are in statistics or
mathematics, many others are in the natural and social sciences and other fields.
In writing this book I have assumed the same diversity of backgrounds; the only
common factor I feel I can take for granted is some previous course in statistics.
The chapters are for the most part organized so that the basic methods and worked
examples come first, with generalizations and key derivations following for those
interested.
A basic one-semester course in sampling can consist of Chapters 1 through 8
and 11 through 13 or 14, with one or more topics from the remainder of the book
added, depending on time and interest. For a graduate class in which many of the
students are interested in the special topics of the last three parts of the book, the
instructor may wish to cover the basic ideas and methods of the first three parts
quite quickly, drawing on them for background later, and spend most of the time
on the second half of the book.
I would like to give my thanks to the many people who have influenced and
enriched the contents of this book through conversations, joint work, and other
interactions on sampling and statistics. In particular, I would like to express appre-
ciation to Fred Ramsey, P. X. Quang, Dana Thomas, and Lyle Calvin. Also, I
am grateful to Lyman McDonald, David Siegmund, Richard Cormack, Stephen
preface to the first edition xxi

Buckland, Bryan Manly, Scott Overton, and Tore Schweder for enlightening con-
versations on statistical sampling methods. I would like to thank my colleagues at
Auckland—George Seber, Alastair Scott, Chris Wild, Chris Triggs, Alan Lee, Peter
Danaher, and Ross Ihaka—for the benefits of our collaborations, discussions, and
daily interactions through which my awareness of relevant and interesting issues
in sampling has been increased. I thank my sabbatical hosts at the Institute of
Mathematical Statistics at the University of Copenhagen, where some of the sam-
pling designs of this book were first seen as sketches on napkins in the lunch
room: Søren Johansen, Tue Tjur, Hans Brøns, Martin Jacobsen, Inge Henningsen,
Søren Tolver Jensen, and Steen Andersson. Among the many friends and asso-
ciates around Alaska who have shared their experiences and ideas on sampling
to the benefit of this book are Pat Holmes, Peter Jackson, Jerry McCrary, Jack
Hodges, Hal Geiger, Dan Reed, Earl Becker, Dave Bernard, Sam Harbo, Linda
Brannian, Allen Bingham, Alan Johnson, Terry Quinn, Bob Fagen, Don Marx, and
Daniel Hawkins. Questions and comments leading to rethinking and rewriting of
sampling topics have been contributed by many students, to each of whom I offer
my thanks and among whom I would particularly like to mention Cheang Wai
Kwong, Steve Fleischman, Ed Berg, and Heather McIntyre.
I would like to give a special thanks to my editor, Kate Roach, at John Wiley
& Sons for her encouragement and enthusiasm. Research support provided by two
grants from the National Science Foundation (DMS-8705812, supported by
the Probability and Statistics Program and DMS-9016708, jointly supported by
the Probability and Statistics Program and the Environmental Biology Division)
resulted in a better book than would have otherwise been possible. I wish to thank
Mary for, among many other things, her supportive sense of humor; when on a
trip through Norway I could not find a certain guide book after ransacking the
luggage jumble from one end of our vehicle to the other, she reminded me to
“use adaptive sampling” and, starting with the location of another book randomly
discovered amidst the chaos, soon produced the wanted volume. Finally, I thank
Jonathan, Lynn, Daniel, and Christopher for an environment of enthusiasm and
innovativeness providing inspiration all along the way.

Steven K. Thompson
Auckland, New Zealand
CHAPTER 1

Introduction

Sampling consists of selecting some part of a population to observe so that one

may estimate something about the whole population. Thus, to estimate the amount
of lichen available as food for caribou in Alaska, a biologist collects lichen from
selected small plots within the study area. Based on the dry weight of these speci-
mens, the available biomass for the whole region is estimated. Similarly, to estimate
the amount of recoverable oil in a region, a few (highly expensive) sample holes
are drilled. The situation is similar in a national opinion survey, in which only a
sample of the people in the population is contacted, and the opinions in the sam-
ple are used to estimate the proportions with the various opinions in the whole
population. To estimate the prevalence of a rare disease, the sample might consist
of a number of medical institutions, each of which has records of patients treated.
To estimate the abundance of a rare and endangered bird species, the abundance
of birds in the population is estimated based on the pattern of detections from a
sample of sites in the study region. In a study of risk behaviors associated with the
transmission of the human immunodeficiency virus (HIV), a sample of injecting
drug users is obtained by following social links from one member of the population
to another.
Some obvious questions for such studies are how best to obtain the sample and
make the observations and, once the sample data are in hand, how best to use them
to estimate the characteristic of the whole population. Obtaining the observations
involves questions of sample size, how to select the sample, what observational
methods to use, and what measurements to record. Getting good estimates with
observations means picking out the relevant aspects of the data, deciding whether
to use auxiliary information in estimation, and choosing the form of the estimator.
Sampling is usually distinguished from the closely related field of experimental
design, in that in experiments one deliberately perturbs some part of a population
in order to see what the effect of that action is. In sampling, more often one likes
to find out what the population is like without perturbing or disturbing it. Thus,
one hopes that the wording of a questionnaire will not influence the respondents’

Sampling, Third Edition. Steven K. Thompson.

© 2012 John Wiley & Sons, Inc. Published 2012 by John Wiley & Sons, Inc.

1
Other documents randomly have
different content
 “Pass in,” said the officer, unbolting
the door and dragging it open.
As soon as the two had stepped
over the sill, the door was slammed
shut behind them, and Bob heard the
great bolts shot into place—and
shuddered in spite of himself. On each
side of him were smooth, solid walls
of rock: ahead of him stretched a
dusky corridor dimly lighted with
electric fireflies suspended here and
there. The dull rumble of distant
machinery came to his ears; the faint
smell of smoke and sulphurous fumes
greeted him.
“Fitz?” the lad said to his comrade,
who stood silent at his side.
The goblin simply gave the speaker
a look in reply.
“Fitz,” Bob continued, “what’s the
meaning of all this talk about my
going through the factories? What’s the matter, anyhow?”
“Nothing—nothing!” Fitz murmured hoarsely, shiftily gazing here
and there.
“Yes, there is,” the boy insisted. “Why do you all emphasize the
word ‘through’?”
“Why—why,” Fitz stammered, rubbing his nose and blinking his
pop eyes, “we thought maybe you didn’t mean that you desired to
go through the factories; thought maybe you meant you desired to
go partly through only—just wanted to see some of the things.”
“No,” Bob hastily made reply, “I want to go through; I want to see
everything. Understand?”
 Fitz nodded.
“Well, come on, then,” he said; “we’ve got to be moving.”
As they went along the corridor, Bob became aware of doors
ahead opening to right and left. He saw the flash of flames and
heard the whirr of wheels and the hub-bub of hammers.
“This room to the right,” said Fitz Mee, “is the machine-shop; that
on the left is the forging-room.”
They visited each in turn, and the lad was delighted with all he
saw.
“He! he!” he laughed when they were again out in the corridor
and free from the thunder and crash and din that had almost
deafened them. “The idea, Fitz, of me not wanting to go through
your factories; of not wanting to see everything! You bet I want to
go through! You thought I’d be afraid—that’s what you thought; and
the mayor, too. But I’ll show you; I’m no baby—not much!”
His companion grinned impishly, but made no reply.
The next place they entered was the great moulding-room. Open
cupolas were pouring forth white-hot streams of molten metal,
which half-nude and sweaty, grimy goblins were catching in ladles
and bearing here and there. The temperature of the room was
almost unbearable; the atmosphere was poisonous with sulphurous
gases. Bob crossed the threshold and stopped.
“Come on,” commanded his companion; “we must hurry along, or
we won’t get through to-day.”
“I—I don’t believe I care to go through here,” Bob said
hesitatingly.
“Why?” Fitz Mee jerked out.
“It’s so awful hot and smelly,” the boy explained; “and I’m—I’m a
little afraid of all that hot metal.”
“No matter; you must go through here.”
 “I must?” Bob cried indignantly.
“Certainly. You said you’d be pleased to go through our factories;
so now you must go through—through every apartment. Boys in
Goblinville, you know, must do what pleases ’em.”
“But it doesn’t please me to go through this fiery furnace, Fitz.”
“Well, boys’re not allowed to change their minds every few
minutes in Goblinville. Come on.”
“I won’t!” Bob said obstinately.
“You’ll get into trouble, Bob.”
“I don’t care.”
“And you’ll get me into trouble.”
“You into trouble? How?”
“You heard what the mayor said, didn’t you?”
“Y-e-s.”
“Well?”
“Well, I’ll go through for your sake, Fitz; but I don’t want to. It is a
fool law or custom—or whatever it is—that won’t let a fellow change
his mind once in a while, when he feels like it! A great way that is to
let a boy do what he pleases! But lead on.”
They sauntered through the moulding—room, Bob trembling and
dodging and blinking, and out into the corridor again.
“Mercy!” the urchin exclaimed, inhaling a deep breath of relief. “I
don’t want any more of that! I’m all in a sweat and a tremble; I was
afraid all the time some of that hot metal would splash on me.”
“It does splash on the workers at times,” Fitz Mee observed
quietly.
Not heeding his companion’s remark, Bob continued: “And my
lungs feel all stuffy. I couldn’t stand such a hot and smelly place
more than a few minutes.”
 “How do you suppose the moulders stand it for ten hours a day?”
Fitz asked.
“I don’t see how they do—and I don’t see why they do,” the boy
replied.
“You don’t see why they do?”
“No, I don’t.”
“For the same reason workmen stand disagreeable and dangerous
kinds of work in your country, Bob; to earn a living.”
“I wouldn’t do it,” the boy declared loftily.
“You might have to, were you a grown man or goblin.”
“Well, I wouldn’t. My papa doesn’t have to do anything of the
kind.”
“Your father’s a physician, isn’t he?”
“Yes.”
“Well, doesn’t he miss meals, and lose sleep, and worry over his
patients, and work sometimes for weeks at a time without rest or
peace of mind?”
“Yes, he does.”
“But you’d rather do that than be a common laborer for eight or
ten hours a day, would you?”
“I—I don’t know; I’d rather just be a boy and have fun all the
time. And I guess I’ve seen enough of your factories, Fitz; I want to
get out into the fresh air and sunshine again.”
“You must go on through,” the goblin answered, quietly but
positively.
“Well, have we seen nearly all there is to see?”
“No, we’ve just begun; we haven’t seen one-tenth part yet.”
 “Oh, dear!” Bob groaned. “I never can stand it, Fitz; it’ll take us all
day.”
“Yes,” the goblin nodded.
“Well, I tell you I can’t stand it.”
“But you must; it was your choice.”
“Choice!” angrily. “I didn’t know What it would be like.”
“You shouldn’t have chosen so rashly. Come on.”
Bob demurred and pleaded, and whimpered a little, it must be
confessed; but his guide was inexorable.
It is not necessary to enter into details in regard to all the boy
saw, experienced and learned. Let it suffice to say that at three
o’clock that afternoon he was completely worn out with strenuous
sight-seeing. The grating, rumbling, thundering sounds had made
his head ache; the sights and smells had made his heart sick. He
had seen goblins, goblins, goblins—goblins sooty and grimed,
goblins wizened and old before their time; goblins grinding out their
lives in the cutlery factory; goblins inhaling poisonous fumes in the
chemical works; goblins, like beasts of burden, staggering under
heavy loads; goblins doing this thing, that thing and the other thing,
that played havoc with their health and shortened their lives. And he
was disgusted—nauseated with it all!
“Oh, Fitz!” he groaned. “I can’t go another step; I can’t stand it to
see any more! I thought it would be pleasant; but—oh, dear!”
“Sit down here and rest a minute,” Fitz Mee said, not unkindly,
indicating a rough bench against the wall of the corridor. “Now, why
can’t you bear to see any more?”
“Oh, it’s so awful!” the boy moaned. “I can’t bear to see ’em
toiling and suffering, to see ’em so dirty and wretched.”
The goblin laughed outright.
“Bob, you’re a precious donkey!” he cried. “True, the workers in
the factories toil hard at dirty work—work that shortens their lives in
some cases; but they’re inured to it, and they don’t mind it as much
as you think. And what would you? All labor is hard, if one but thinks
so; there are no soft snaps, if one does his duty. It’s the way of the
goblin world, and it’s the way of the human world. All must labor, all
must suffer more or less; there’s no escape for the highest or the
lowest. And work has its compensation, has its reward; it—”
“Oh, shut up!” the lad muttered petulantly. “I don’t want to hear
any more. You talk just like my papa does. I wish I’d never been
born, if I’ve got to grow up and work. So there!”
“You’ll never grow up, if you stay in Goblinville, Bob,” Fitz Mee said
softly; but his pop eyes were twinkling humorously. “And you won’t
have to work—not much, anyhow.”
Bob sat soberly silent; evidently he was doing some deep thinking.
The goblin went on: “If you’re rested now, we’ll resume our sight-
seeing.”
“I don’t want to see any more,” the lad grunted pugnaciously;
“and I’m not going to, either.”
“Yes, come on.”
“I won’t do it.”
“Please do, Bob.”
“I won’t, I say.”
“You’ll get us both into trouble.”
“I don’t care if I do.”
“They’ll send us to prison.”
“What!”
“They will.”
“Who will?”
“The mayor and his officers.”
 “Send us both?”
“Yes.”
“Well,” bristling, “I guess they won’t send me—the old meddlers!
They won’t dare to; I’m not a citizen of this country.”
“That won’t make any difference, Bob!”
“It will too. If they send me to prison, the people of my country
will come over here and—and lick ’em out of their boots. Now!”
Fitz Mee bent double and stamped about the floor, laughing till the
tears ran down his fat cheeks. But suddenly he sobered and said:
“Come on, Bob; you’ve got to.”
“I won’t!” the boy declared perversely. “I don’t have to.”
The goblin made no further plea; but placing a silver whistle to his
lips blew a sharp blast. In answer, a squad of officers stepped from
the shadows.
“What’s wanted, Fitz Mee?” said the leader.
“This boy flatly refuses to obey the law, to go on through the
factories, as he stated would please him.”
“Boy, is this true?” demanded the officer.
“Yes, it is,” Bob confessed fearlessly, shamelessly.
“Fitz Mee, he confesses,” muttered the officer. “What would you
have me do?”
“Take him and carry him through,” Fitz Mee said icily.
“Very well,” answered the officer. “But if we do that we take the
case out of your hands, Fitz Mee. And in order to make a satisfactory
report to the mayor, we’ll have to carry him through all the factories
—those he has already visited as well as those he has not.”
“Yes, that’s true,” Fitz nodded.
“What’s that?” Bob cried, keenly concerned.
 The officer gravely repeated his statement.
“Oh, nonsense!” the boy exclaimed. “You
fellows go away and quit bothering me. I never
saw such a country! A fine place for a boy to do
as he pleases, surely! Come on, Fitz.”
All the goblins laughed heartily, and Bob
disrespectfully made faces at them, to their
increased amusement.
When the two comrades had made their round
of the factories, and were out in the fresh air
again, the boy murmured meekly, a sob in his
throat:
“Fitz, I’m tired—I’m sick of it all! I wish I hadn’t
come here, I—I wish I was back home again.”
“What!” his companion cried in assumed
surprise.
“I do!”
“Back home, and be compelled to obey your elders—your parents
and your teachers?” Fitz Mee said, grinning and winking impishly.
“Well,”—pettishly,—“it wouldn’t be any worse than being
compelled to obey a lot of fool officers, anyhow.”
“You’re just compelled to do what pleases you, just as I told you,”
Fitz Mee explained smoothly.
“Oh, do shut up!” the lad pouted.
“You’re out of sorts,” the goblin giggled; “you’re hungry—you need
some food tablets.”
“Bah!” Bob gagged. “Pills! I can’t swallow any more of ’em—I just
can’t! Oh, I wish I had a good supper like mother cooks!”
Fitz Mee threw himself prone and kicked and pounded the earth,
laughing and whooping boisterously; and Bob stood and stared at
him, in silent disapproval and disgust.
 CHAPTER XII
BEFORE THE MAYOR OF GOBLINLAND

s the days passed Bob became more and more

disgruntled, more and more dissatisfied with
things in Goblinville. The bare thought of food-
tablets and drink-pellets disgusted and
nauseated him; and he could hardly swallow
them at all. The young goblins would not,
could not, play the games he liked to play.
They were too small for one reason; and, then, as it did not please
them to do so, they were not permitted to do so. And the boy was
without youthful companionship. The only associates he had were
his faithful companion Fitz Mee and the officers of the town, who
were always at his elbow to see that he did what pleased him. This
constant espionage became simply unbearable; and the lad grew
peevish, gloomy; desperate. At last he broke down and tearfully
confessed to his comrade:
“Fitz, I want to go back home; I do—I do! I can’t stand it here any
longer. It isn’t at all what I thought it would be like; and I’m
homesick!”
Fitz Mee did not laugh; he did not smile, even. On the contrary he
looked very grave—and a little sad.
“So you’re homesick, Bob—eh?” he said.
“Yes, I am, Fitz.”
“And you desire to go home?”
“Uh-huh.”
“You don’t like things here in Goblinville?”
 “No, I don’t.”
“What is it you object to?”
“Oh, everything!”
“But especially?”
“Well, the—the pills, I guess.”
“Oh!” joyfully. “Is that all, Bob? We can fix that all right. I’ll get a
special permit from the mayor—he’s a political friend of mine,—to let
me prepare you food like you’ve been accustomed to. Then you’ll be
as happy as a clam, won’t you?”
“I—I don’t hardly know, Fitz; no, I don’t think I will.”
“What!”
“Uk-uh.”
“Well, what else is wrong, then?”
The goblin’s pop eyes were dancing with mischief.
“I don’t like to be compelled to do what pleases me,” Bob
confessed shamefacedly.
“Ho, ho!” laughed Fitz Mee.
“Oh, you can laugh!” the boy cried, in weak irritation. “But I
don’t!”
“You said it would just suit you, Bob—before you came here,” Fitz
chuckled hoarsely, holding his sides and rocking to and fro.
“I know I did; but I’d never tried it.”
“And you don’t like it?”
“No, indeed,” Bob answered very earnestly.
“And you’re homesick, and want to go home?”
The boy nodded, his eyes downcast.
“All the goblins’ll laugh at you, if you go to leave Goblinville.”
 “Well, let ’em; I don’t care.”
“And your people and your schoolmates will laugh at you, when
you return home.”
Bob was silent, deeply pondering.
“Don’t you care?” Fitz Mee asked, cackling explosively.
“Yes, I do! But I’ve got to go, anyhow; I’ll die here.”
“Oh, no, you won’t, Bob,” said the goblin, teasingly.
“I will, too,” said Bob, desperately in earnest; “I know.”
“You’ll have to go to school, if you return home.”
“I don’t mind that; I’ll have other boys to play with, anyhow.”
“Yes, but you’ll have to obey the teacher.”
“I know.”
“And you’ll have to do what pleases your parents.”
“I know that, too.”
“And you won’t be permitted to do what pleases yourself.”
“I know; I’ve thought it all over, Fitz.”
“And yet you wish to return home?”
“Yes, I do.”
Fitz Mee laughed gleefully, uproariously, irrationally, laughed till
the tears coursed down his cheeks and his fat features were all a-
quiver.
“Ho, ho!” he gasped at last. “Roberty-Boberty, you’re not the same
boy you were, not at all; you’re not half as high and mighty. What’s
come over you, hey?”
“I’ve—I’ve learned something, I—I guess, Fitz.”
“Oh, you have!”
“Uh-huh.”
 “What?”
“I’m not going to say,” replied Bob, grinning sheepishly; “but I
think I know what you brought me to Goblinland for.”
“What for?”
“W-e-ll, to—to teach me what I’ve learned. Didn’t you?”
“I’m not going to say,” mimicked the goblin.
Then both tittered.
“And you’re bound to go back home, Bob?” Fitz pursued.
The boy nodded.
 “If you’re rested now, we’ll resume our sight-seeing.” (See page
168.)

“You’re a pretty looking thing to go back to Yankeeland—a little

mite of a human like you!” sneeringly.
“Oh, Fitz!” the lad wailed. “Can’t I be made a real boy again?”
The goblin impressively shook his head.
“I don’t know,” he said slowly. “You see you’ve taken so many
gob-tabs it’s very doubtful whether you can be changed back into a
boy at all.”
“Oh, Fitz, don’t say that!”—greatly distressed.
“Of course, if you were put on human diet for a long time, you
might come out all right,”—reflectively.
“But can’t I take something that will change me quick—right
away?”
Again the goblin shook his head.
“I doubt it,” he murmured. “Giant-tabs would make a giant of you;
and you don’t want to be a giant.”
“No, I don’t.”
“Well, I guess, then, if you want to go back home right away,
you’ll have to go just as you are.”
“Oh, Fitz!” almost blubbering. “I don’t want to go back home this
way; I just can’t! Can’t you give me something that will—will stretch
me and swell me to boy size—just to boy size? Can’t you—can’t
you?”
“I don’t know,”—with a gloomy shake of the head; “I never heard
of such a drug or chemical, but it’s barely possible our chemists may
know of something of the kind. I’ll see about it. But here’s a
difficulty.”
“What? What, Fitz?”
“Why, as you know, there’s no means of getting out of Goblinland
except by balloon; and I doubt if my balloon will carry you at full and
normal weight.”
“But can’t you get a bigger one?”
“I might have one made; I don’t—”
“Oh, no—no, Fitz!” the boy interrupted frantically. “Don’t think of
doing that; I can’t wait. Can’t you borrow a bigger one?”
 “There are no bigger ones, except the mayor’s state balloon. It
has two feather beds lashed together for a bag, and a very large
car.”
“Can’t you get it—can’t you get it, Fitz?”
“I don’t know, indeed. Then, here’s another difficulty, Bob, and a
greater one to my mind.”
“Oh, Fitz!” the boy moaned, wringing his hands. “You don’t mean
it!”
“Yes, I do,” said the goblin, nodding gravely; but his twinkling pop
eyes belied his words. “You see, Bob, you’re the first human being
that has ever come to Goblinland. Now, the secrets of the country—
including the secret of its whereabouts, have always been carefully
guarded. I don’t know what his honor, the mayor, will say about
letting you go.”
“I won’t tell anything, Fitz, I won’t—I won’t!”
“Not a thing?” questioned Fitz Mee.
“No, sir—not a thing.”
“W-e-ll, I—I don’t know. What will you do, Bob, if the mayor won’t
let you go back home?”
“I’ll just die—that’s what!”
The goblin slapped his thin thighs and laughed and whooped, and
laughed some more.
Out of patience, the lad screamed: “Laugh! Laugh till you burst,
you old Convulsions! You old Spasms! You old Hysterics! Yeah!
Yeah!”
And Fitz Mee did laugh—till he was entirely out of breath and
panting and wheezing like a bellows. When at last he had regained
control of himself, he whispered brokenly:
“Bob, we’ll—we’ll go and see—the mayor.”
And they caught up their caps and were off.
 “So you wish to go home, boy—eh?” said the mayor, the august
ruler of Goblinville and all adjacent territory, as soon as the two were
ushered into his presence.
“Yes, sir,” Bob answered humbly. Then, with boyish inquisitiveness:
“But how did you know it?”
“Never mind,” was the gruff reply. “It will please you to return
home will it?”
“Yes, sir, indeed it will.”
“Then you must go. Be off at once.”
“But—but—” Bob began.
“I’ll fix all that,” his honor interrupted, quickly divining what the
boy meant to say. “I’m as anxious to be rid of you as you are to be
gone. You’ve stirred up a pretty rumpus here—you have. You’re the
first human boy that ever came into my domain; and you’ll be the
last. But I trust your experience has done you good—eh?”
Bob nodded.
“Very well, then. Sign this pledge that you won’t reveal what
you’ve seen and learned, and that you’ll take the lesson to heart.”
Bob gladly signed the pledge.
“Now,” continued the mayor, his eyes snapping humorously, “these
are the conditions under which you must leave my domain: I’ll call in
the chemists and have them restore you to normal size; I’ve already
communicated with them, and they assure me they can do it. Then
I’ll let the honorable and worthy Fitz Mee take my state balloon and
carry you back to Yankeeland. You will set out this afternoon at one
o’clock. But one other thing I exact: you must bear nothing away
with you that you did not bring here with you.” And the mayor gave
the boy a keen, meaningful look that the urchin could not interpret.
 The chemists came in, three aged and bewhiskered goblins
wearing long, black robes and silk skull caps.
“My good chemists,” said the mayor, “are you ready for the
experiment?”
“All ready, your honor,” the eldest of the three made answer,
bowing profoundly.
“To work, then,” the mayor commanded.
The younger two advanced and caught and held Bob’s hands,
their fingers upon his pulse. The eldest produced a tiny phial of
thick, opalescent liquid.
“Put out your tongue,” he said to the boy.
The lad unhesitatingly obeyed, and the aged and trembling
chemist let a drop of the viscid liquid fall upon the tip of the
youngster’s quivering organ of speech.
 The effect was instantaneous and startling, if not marvelous. Bob
let out a mad bellow of pain, shaking his head and writhing and
drooling. The mayor changed countenance and deprecatingly shook
his head. Fitz Mee groaned aloud.
“Draw in your tongue and shut your mouth and swallow!” the
three savants simultaneously yelled at the boy.
Bob reluctantly did as he was told; and immediately,
instantaneously he was restored to normal size.
“Whoopee!” shouted the chemists, embracing one another and
indulging in mad capers and other manifestations of insane joy. “A
success! A complete success!”
“Thank goodness!” murmured Fitz Mee. “A success!”
“Yes,” the mayor muttered drily, grimly, “a remarkable success—a
too remarkable success! My good chemists, destroy what you have
left of that stuff, and make no more at your peril. I’m not going to
have any more boys manufactured in this country—a noisy,
disturbing lot! You hear me!” Then to Fitz Mee: “You take your
departure from the public square at one o’clock, remember. The
state balloon will be there in readiness. You’re excused.”
When the two comrades were again at Fitz Mee’s residence, Bob
remarked ingenuously:
“Fitz, while you’re getting ready I’m going to gather up some of
the gold nuggets I saw on the shore of the brook.”
“Better not,” Fitz replied, without looking up from his work.
“Why?”
“I wouldn’t, if I were you—that’s all.”
“Well, why?”
“They’re not yours.”
“I know. But you goblins make no use of them; and it wouldn’t be
wrong—wouldn’t be stealing, would it?”
 “No,” Fitz Mee mumbled, “it wouldn’t be robbery, exactly. But you
heard what the mayor said.”
“What about?”
“That you weren’t to take anything away with you that you didn’t
bring here with you.”
“Yes, I heard him. Is that what he meant?”
“To be sure.”
“Well, why does he object to my taking a few old nuggets of gold
that none of you will use?” said Bob peevishly.
“For this reason, Bob: you take that gold back to Yankeeland, and
tell where you got it—”
“But I won’t tell where I got it,” the lad interrupted.
Unheeding, the goblin continued: “And your money-mad people
will search out our country and conquer and ruin us.”
“Oh, pshaw, Fitz!”
“What I say is true, Bob.”
But Bob was neither convinced nor satisfied, and he resolved to
have the nuggets at all hazard. Where was the harm? The gold was
of no value to the goblins; it would be of great value to him. And he
wouldn’t say a word about where he got it—indeed he wouldn’t. He
would take it; and no one would be the wiser or the poorer. So,
while his comrade was busy at other things, he slipped out to the
brookside and filled his pockets.
One o’clock came, the time of departure, and all Goblinville,
including the mayor and his officers, was out to see the aëronauts
off upon their long voyage. The mayor shook hands with the two
and wished them God-speed and the populace gave them three
hearty cheers.
Then the anchor was weighed, and they were off. Slowly and
majestically the great state balloon began to ascend. But when it
had risen a hundred feet, Bob, looking over the side of the car,
became aware of a disturbance in the crowd beneath. He saw
goblins excitedly running this way and that and a number of officers
trundling a big black object on wheels across the public square.
“What’s the meaning of the rumpus, Fitz?” the lad cried to his
companion. “What’s that the officers have?”
“Why,” Fitz gasped, taking a hurried look beneath, “the officers are
running out the dynamite gun!”
“And they’re training it upon our balloon—upon us!” Bob
whispered hoarsely, his soul a prey to guilty fear. “What—what can it
mean, Fitz?”
Then arose the voice of the mayor, bellowing:
“Fitz Mee, descend! Come back! That boy can’t leave Goblinland
with his pockets full of gold! He has deceived us; he can’t leave
Goblinland at all. Come down; or we’ll send a dynamite shell through
the balloon-bag, and bring you down in a hurry.”
Fitz gave a few strokes to the pump, and the big balloon came to
a stop. Bob sat silent, speechless at the dread result of his rash act.
“You’ve played the mischief—you have, Bob Taylor!” his
companion snarled angrily, reproachfully. “And you’ll spend the
balance of your days in Goblinland—that’s what!”
“Oh, dear!” the boy found voice to moan. “Oh, dear!”
“Hello!” Fitz called over the side of the car. “Hello, your honor!”
“Hello!” answered the mayor.
“If I’ll make the boy throw the gold down to you, will that satisfy
you?”
“No, it won’t!” came the hoarse and determined reply. “Bring the
young scamp back! He shall stay in Goblinville!”
“I guess I won’t!” Bob shouted, desperation spurring his courage.
And he sprang to the air-tank and opened the cock. The balloon
began to rise swiftly.
“Oh, Bob—Bob!” Fitz Mee groaned. “What have you done? We’ll
both be killed!”
“Boom!” went the dynamite gun; and a shell tore through the
balloon-bag, rending it asunder and sending goose feathers
fluttering in all directions.
The car began to drop like a plummet. Its occupants let it out
shrill screeches of terror. Then came the proverbial dull, sickening
thud! Bob felt the empty balloon—bag fall over him and envelop
him; and then he lost consciousness.
“Bob, crawl out of there.”

“Fitz! Fitz!” the boy cried, disentangling himself and struggling to

his feet.
“Fits!” laughed a big manly voice. “Yes, I guess you’ve got ’em,
Bob; and you’ve rolled out of bed in one, and dragged the covers
with you.”
Bob blinked and rubbed his sleepy eyes. There stood his father in
the doorway, grinning broadly;
“Hustle into your clothes, laddie,” he said; “breakfast’s ready.”
*** END OF THE PROJECT GUTENBERG EBOOK THE LITTLE GREEN
GOBLIN ***

Updated editions will replace the previous one—the old editions will
be renamed.

Creating the works from print editions not protected by U.S.

copyright law means that no one owns a United States copyright in
these works, so the Foundation (and you!) can copy and distribute it
in the United States without permission and without paying
copyright royalties. Special rules, set forth in the General Terms of
Use part of this license, apply to copying and distributing Project
Gutenberg™ electronic works to protect the PROJECT GUTENBERG™
concept and trademark. Project Gutenberg is a registered trademark,
and may not be used if you charge for an eBook, except by following
the terms of the trademark license, including paying royalties for use
of the Project Gutenberg trademark. If you do not charge anything
for copies of this eBook, complying with the trademark license is
very easy. You may use this eBook for nearly any purpose such as
creation of derivative works, reports, performances and research.
Project Gutenberg eBooks may be modified and printed and given
away—you may do practically ANYTHING in the United States with
eBooks not protected by U.S. copyright law. Redistribution is subject
to the trademark license, especially commercial redistribution.

START: FULL LICENSE

THE FULL PROJECT GUTENBERG LICENSE
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!

ebookultra.com