Theory Construction and Model Building Skills A

Practical Guide for Social Scientists 2nd

Edition James Jaccard install download

https://textbookfull.com/product/theory-construction-and-model-
building-skills-a-practical-guide-for-social-scientists-2nd-
edition-james-jaccard/

Download more ebook instantly today - get yours now at textbookfull.com

We believe these products will be a great fit for you. Click
the link to download now, or visit textbookfull.com
to discover even more!

Social Skills Handbook Practical Activities for Social

Communication 2nd Edition Sue Hutchings

https://textbookfull.com/product/social-skills-handbook-
practical-activities-for-social-communication-2nd-edition-sue-
hutchings/

Model Predictive Control: Theory, Computation, and

Design, 2nd Edition Rawlings James B.

https://textbookfull.com/product/model-predictive-control-theory-
computation-and-design-2nd-edition-rawlings-james-b/

Building a Roll-Off Roof or Dome Observatory: A

Complete Guide for Design and Construction 2nd Edition
John Stephen Hicks (Auth.)

https://textbookfull.com/product/building-a-roll-off-roof-or-
dome-observatory-a-complete-guide-for-design-and-
construction-2nd-edition-john-stephen-hicks-auth/

Practical Statistics for Data Scientists 2nd Edition

Peter Bruce

https://textbookfull.com/product/practical-statistics-for-data-
scientists-2nd-edition-peter-bruce/
Successful Academic Writing A Complete Guide for Social
and Behavioral Scientists 1st Edition Anneliese A.
Singh

https://textbookfull.com/product/successful-academic-writing-a-
complete-guide-for-social-and-behavioral-scientists-1st-edition-
anneliese-a-singh/

A Practical Guide to Construction Adjudication 1st

Edition Pickavance

https://textbookfull.com/product/a-practical-guide-to-
construction-adjudication-1st-edition-pickavance/

Improvised Theatre and the Autism Spectrum A Practical

Guide to Teaching Social Connection and Communication
Skills 1st Edition Kramer

https://textbookfull.com/product/improvised-theatre-and-the-
autism-spectrum-a-practical-guide-to-teaching-social-connection-
and-communication-skills-1st-edition-kramer/

The Naked Consultation A Practical Guide to Primary

Care Consultation Skills 2nd Edition Liz Moulton

https://textbookfull.com/product/the-naked-consultation-a-
practical-guide-to-primary-care-consultation-skills-2nd-edition-
liz-moulton/

Rotor and Structural Dynamics of Turbomachinery A

Practical Guide for Engineers and Scientists 1st
Edition Raj Subbiah

https://textbookfull.com/product/rotor-and-structural-dynamics-
of-turbomachinery-a-practical-guide-for-engineers-and-
scientists-1st-edition-raj-subbiah/
Theory Construction and Model-Building Skills
Methodology in the Social Sciences
David A. Kenny, Founding Editor
Todd D. Little, Series Editor
www.guilford.com/MSS

This series provides applied researchers and students with analysis and research design books that
emphasize the use of methods to answer research questions. Rather than emphasizing statistical
theory, each volume in the series illustrates when a technique should (and should not) be used and
how the output from available software programs should (and should not) be interpreted. Common
pitfalls as well as areas of further development are clearly articulated.

REC ENT VOL UME S

PRINCIPLES AND PRACTICE OF STRUCTURAL EQUATION MODELING, FOURTH EDITION

Rex B. Kline

HYPOTHESIS TESTING AND MODEL SELECTION IN THE SOCIAL SCIENCES

David L. Weakliem

REGRESSION ANALYSIS AND LINEAR MODELS: CONCEPTS, APPLICATIONS,

AND IMPLEMENTATION
Richard B. Darlington and Andrew F. Hayes

GROWTH MODELING: STRUCTURAL EQUATION

AND MULTILEVEL MODELING APPROACHES
Kevin J. Grimm, Nilam Ram, and Ryne Estabrook

PSYCHOMETRIC METHODS: THEORY INTO PRACTICE

Larry R. Price

INTRODUCTION TO MEDIATION, MODERATION, AND CONDITIONAL

PROCESS ANALYSIS: A REGRESSION-BASED APPROACH, SECOND EDITION
Andrew F. Hayes

MEASUREMENT THEORY AND APPLICATIONS FOR THE SOCIAL SCIENCES

Deborah L. Bandalos

CONDUCTING PERSONAL NETWORK RESEARCH: A PRACTICAL GUIDE

Christopher McCarty, Miranda J. Lubbers, Raffaele Vacca, and José Luis Molina

QUASI-EXPERIMENTATION: A GUIDE TO DESIGN AND ANALYSIS

Charles S. Reichardt

THEORY CONSTRUCTION AND MODEL-BUILDING SKILLS:

A PRACTICAL GUIDE FOR SOCIAL SCIENTISTS, SECOND EDITION
James Jaccard and Jacob Jacoby
Theory Construction
and Model-Building Skills
A Practical Guide for Social Scientists
SECOND EDITION

James Jaccard
Jacob Jacoby

Series Editor’s Note by Todd D. Little

THE GUILFORD PRESS

New York  London
Copyright © 2020 The Guilford Press
A Division of Guilford Publications, Inc.
370 Seventh Avenue, Suite 1200, New York, NY 10001
www.guilford.com

All rights reserved

No part of this book may be reproduced, translated, stored in a retrieval system,

or transmitted, in any form or by any means, electronic, mechanical,
photocopying, microfilming, recording, or otherwise, without written permission
from the publisher.

Printed in the United States of America

This book is printed on acid-free paper.

Last digit is print number: 9 8 7 6 5 4 3 2 1

Library of Congress Cataloging-in-Publication Data

Names: Jaccard, James, author. | Jacoby, Jacob, author.
Title: Theory construction and model-building skills : a practical guide
for social scientists / James Jaccard, Jacob Jacoby.
Description: Second edition. | New York : The Guilford Press, [2020] |
Series: Methodology in the social sciences | Includes bibliographical
references and index.
Identifiers: LCCN 2019030892 | ISBN 9781462542437 (paperback) |
ISBN 9781462542444 (hardcover)
Subjects: LCSH: Social sciences—Research—Methodology. | Theory
(Philosophy)
Classification: LCC H62 .J29 2020 | DDC 300.72—dc23
LC record available at https://lccn.loc.gov/2019030892
 For Marty Fishbein—
a brilliant and inspiring theorist and mentor
—James Jaccard

For Renee and Dana—

in appreciation for the balance and joy they create
—Jacob Jacoby
Series Editor’s Note

Theory is an analyst’s best friend, while data are the fodder of good theory; but which
comes first, the data or the theory? Similarly, some folks never let data get in the way
of good theory and some folks never let theory get in the way of good data; but is it this
either/or taming-­of-­the-­shrew-­like scenario? Constructing a theory is more like crafting
an elegant ensemble of logically connected ideas that depict the world and allow knowl-
edge to leap forward. As Jaccard and Jacoby point out, 90% of our graduate training is
on methods for collecting and techniques for analyzing data and only 10% is spent on
identifying and crafting the ideas into theories that can be tested.
I have always been a student of good theory and this book is a veritable bible on
how to craft testable theories, even before the wonderful enhancements to this second
edition. Enhancements include, for example, what constitutes a theoretical contribution
and how do you craft one? How do you use a logic model to generate ideas and avoid
pitfalls in the theory construction process? How do you use mixed methods and data
mining to craft good theory? How do you test a theory and revise it if need be? Jaccard
and Jacoby answer these questions with practical wisdom born of extensive experience
and uncommon insights.
In addition, the expanded discussion of moderator variables, counterfactual causal-
ity, and their new 10-­step method for generating theory is simply invaluable. Oh, and the
new chapters on measurement are absolutely essential because good theory demands
good measurement. The operationalization of good theory is way too often neglected in
training programs and many seasoned veterans have yet to learn how to do good mea-
surement. Last but not least, they offer insight and wisdom on both interpreting others’
theories as well as clearly expressing your own theories.
Their book is transdisciplinary, with many useful examples spanning fields such
as anthropology, business, communications, education, economics, health, marketing,
organizational studies, political science, psychology, social work, sociology, and so on.
vii
viii Series Editor’s Note

And the companion website is a tremendously useful and helpful resource for students
and instructors alike! I will be teaching a course on model building and theory con-
struction. Jaccard and Jacoby is unequivocally the only choice for such a course.
Sadly, Jacob “Jack” Jacoby has departed our worldly sphere. Brilliant thinkers chal-
lenge our worldview on humanity and they leave indelible marks that shape us and how
we think. Jack was one of our field’s finest.

Todd D. Little
Honeymooning at Casa Montana
Preface

Theory construction is at the heart of the scientific process. The strategies that social
scientists use to generate and develop ideas are important to understand and foster
in young academics and investigators as they prepare for a research-­oriented career.
Although books have been written about theory construction, there are surprisingly few
books on the topic that tackle the problem of teaching students and young professionals,
in a practical and concrete way, how to theorize. Students, especially graduate students,
take one or more courses on research methods and data analysis, but few experience
more than a lecture or two, or read a chapter or two, on theory construction. It is no
wonder that students often are intimidated by the prospect of constructing theories.
This book provides young scientists with tools to assist them in the practical
aspects of theory construction. It is not an academic discussion of theory construction
or the philosophy of science, and we do not delve too deeply into the vast literature on
these topics. Rather, we take a more informal journey through the cognitive heuristics,
tricks of the trade, and ways of thinking that we have found to be useful in developing
theories—­essentially, conceptualizations—­that can advance knowledge in the social
sciences. By taking this journey, we hope to stimulate the thinking and creative pro-
cesses of readers so that they might think about phenomena in new and different ways,
perhaps leading to insights that might not otherwise have resulted. The intent of this
book is to provide a practical, hands-­on, systematic approach to developing theories and
fostering scientific creativity in the conceptual domain. Relative to the majority of books
on theory construction, this book is unique in its focus on the nuts and bolts of building
a theory rather than on an analysis of broad-­based systems of thought.
We have used the book both as a stand-­alone text in a course on theory construction
and as one of several texts in graduate courses on research and research methodology.
In terms of the latter, almost all traditional research methods books include a section
or chapter on the nature of theory and/or theory construction. However, the treatment
ix
x Preface

of theory construction usually is brief and of limited practical value. The present book
is intended to provide the instructor with a useful source for helping students come up
with ideas for research and for fine-­tuning the resulting theories that emerge from such
thinking. It provides more detail and more practical knowledge than what is typical of
chapters in books on research methodology. The social psychologist William McGuire
often lamented about how research training with graduate students focuses at least 90%
on teaching methods to test ideas but no more than 10% on how to get those ideas in
the first place. Despite this difference in emphasis, the process of theory development is
fundamental to successful scientific research. Indeed, many would say that there can be
no theory testing without theory. An objective of this book is to move toward a needed
balance in the emphases given to theory construction and theory testing.
The book can be used in many different disciplines. We draw on examples from the
fields of anthropology, business, communications, education, economics, health, mar-
keting, organizational studies, political science, psychology, social work, and sociology,
to name a few. Some instructors may prefer more detailed examples in their particular
field of study, but we believe that using examples from multiple disciplines helps stu-
dents appreciate the commonalities and value of multidisciplinary perspectives.
The book has several pedagogical features that enhance its use as a textbook and as
a source of learning. First, each chapter includes a section on suggested readings with
commentary, where we direct the reader to key references for further study on the topics
covered in the chapter. Second, each chapter has a list of key terms that highlights the
most important jargon and terminology. Third, each chapter has a set of exercises that
encourages the reader to think about the material that was presented in the chapter. We
include exercises to reinforce concepts and exercises to apply the concepts to problems
of interest. Finally, each chapter has a highlighted box that covers an interesting topic
that applies the concepts covered in the chapter or that shows important uses of them.
We also created a website that contains supplemental materials to support the book (see
the box at the end of the table of contents). The website is intended for use by students,
professors, and professionals alike.

CHANGES IN THE SECOND EDITION

In the first edition, we downplayed issues surrounding data collection and data analy-
sis, preferring to keep discussion at a conceptual level. This orientation still dominates
the current edition, but we felt it important to more fully recognize that theory often
emerges from data collection and data analysis. In the first edition, the emergence of
theory from data was front and center in the chapter on grounded and emergent the-
ory because in qualitative research, emergent theory is prominent. However, the emer-
gence of theory from exploratory data analysis was absent for quantitative research. We
have added a new chapter, “Emergent Theory: Quantitative Approaches” (Chapter 11),
to address this and have retitled our chapter on qualitative approaches (Chapter 10)
“Emergent Theory: Qualitative/Mixed-­Methods Approaches.” The quantitative chapter
on this topic prioritizes novel, exploratory methods of quantitative analysis that can
 Preface xi

help readers generate new theory through data mining. Readers who are less interested
in quantitative research may find the core material in this chapter of lesser relevance.
However, we have kept this material conceptual as opposed to being steeped in statisti-
cal theory and have provided supplemental materials on our website that walk readers
through the execution of the methods on popular software at a more practical level.
Another nod to the fact that data often lead to new theory is the addition of a chap-
ter on theory revision (Chapter 15). When we collect data to test a theory, disconfirma-
tory results can emerge that lead us to revise the theory or abandon it altogether. When
faced with disconfirming or only partially supportive data, one can use critical think-
ing processes to make decisions about whether and how to revise a theory. Chapter 15,
“Theory Revision,” highlights these processes. Theory revision in light of disconfirming
data is as relevant to qualitative researchers as it is to quantitative researchers, so this
chapter should be of interest to all.
Measurement and observation are core to science. When we formulate measures, we
invoke theory to link measures to the underlying construct the measures are assumed
to reflect. When we address measurement error in research, measurement theory also
is front and center. As such, measurement theory is a core part of science. Measurement
is typically viewed as the province of methodology, but we seek to build a case with two
new chapters showing that the practice of measurement is firmly entrenched in theory
and that measurement-­oriented theory construction is essential for the social sciences.
The first chapter on measurement, “General Frameworks” (Chapter 13), emphasizes
the concepts of metrics, reliability, validity, and measurement facets. The second chap-
ter, “Types of Measurement Strategies” (Chapter 14), focuses on self-­reports, observer
reports, and “objective” measures, strategies that form the backbone of social science
research. We provide readers with theory construction principles that guide how one
thinks about and conceptualizes such measures. Readers who want to contribute to
measurement theory will learn useful conceptual tools for doing so. Readers who are
not so inclined will still learn about the importance of measurement theory and how to
apply that theory to the specific research projects they pursue. This will be true for both
qualitative and quantitative researchers because both traditions ultimately use measure-
ment in one form or another.
In the first edition, we wrote each chapter so that it could generally “stand on its
own.” The idea was that if instructors wanted to change the reading order of chapters,
omit certain chapters based on their own or their students’ substantive interests, or
browse different topics rather than read every chapter, the book would be amenable to
these approaches. The new chapters have this same quality.
We also have added new material to most chapters from the first edition. As exam-
ples, in Chapter 3, we added a discussion of what constitutes a theoretical contribution
and what strategies social scientists can use to make theoretical contributions. This
helps orient readers to the remainder of the book. We added a section on conceptual
logic models to the chapter on generating ideas (Chapter 4) because such models are key
to idea refinement. We expanded Chapter 6, on thought experiments, to give a better
appreciation of their role in science and reworked several of the example experiments.
We made clearer how thought experiments can be used both by confirmatory-­oriented
xii Preface

and emergent-­oriented theorists. For the chapter on causal thinking (Chapter 7), we
expanded our discussion of moderator variables, added a brief discussion on counter-
factual causality, and added two ways of generating theory, a 10-­step method and a
“binder” method. In this chapter, we also added a section on common mistakes made
during the theory construction process. We expanded the material on grounded and
emergent theory in Chapter 10 to discuss mixed-­methods approaches in more depth and
to develop the actual thought processes theorists use to construct theories from quali-
tative data. We also expanded Chapter 16, on reading about and writing theories, and
generally updated the book to include practical perspectives that have evolved in theory
construction since the first edition.
Our book should be useful (1) in theory construction courses, (2) in proseminars
for doctoral students to help them develop their thesis research, and (3) as a supplement
to methods courses where instructors can select a subset of chapters for students to read.
Young researchers and professors also should find the book of interest independent of
courses. We hope that even seasoned researchers will walk away from most chapters
with at least one or two new “nuggets of knowledge” they will find useful in their work.
In this sense, our intended audience is broad.
As noted, we have created a web page for the book (see the box at the end of the
table of contents). This contains useful supplemental information for readers as well as
instructional aids for professors.

ACKNOWLEDGMENTS

My dear friend and colleague Jacob Jacoby passed away before writing commenced on
the second edition. Despite this, his presence remains even in the new chapters. Jack
was an important influence on my life both professionally and personally, and I will
always be grateful for the opportunity to have known and learned from him. His legacy
will be with us for years. I miss you, Jack.
As with the first edition, a large number of people contributed in diverse ways to the
development of the second edition. I again would like to thank students and colleagues
who provided feedback on earlier drafts, including David Brinberg, Department of Mar-
keting, Virginia Polytechnic University; Miriam Brinberg, Department of Human Devel-
opment, Pennsylvania State University; Wendy J. Coster, Department of Occupational
Therapy, Boston University; Cynthia G. S. Franklin, Steve Hicks School of Social Work,
University of Texas, Austin; Liliana Goldin, Silver School of Social Work, New York
University; Guillermo Grenier, Department of Global and Sociocultural Studies, Florida
International University; Sean Patrick Kelly, School of Education, University of Pitts-
burgh; Hailin Qu, Spears School of Business, Oklahoma State University; Rick Sholette,
Director, Paraclete Ministries; Michael Slater, School of Communications, Ohio State
University; and Weiwu Zhang, Department of Public Relations, Texas Tech University.
At The Guilford Press, C. Deborah Laughton, as always, was insightful and supportive
in her role as editor. She once again improved the book immensely. It is rare one gets the
opportunity for a “do-­over” of a project, and I am grateful to C. Deborah for giving me
 Preface xiii

the opportunity. Although all of these individuals contributed significantly to the book,
I alone am responsible for any of its shortcomings.
I dedicated the first edition to Marty Fishbein and I do so again here. Marty was an
amazing scientist, teacher, and professor who positively impacted my life and those of
his many students. It is rare that one has the opportunity to study under and work with
a true genius. Jack, I am sure, would once again dedicate the book to his spouse, Renee,
and his daughter, Dana, two incredible women. My spouse, Liliana, again contributed
to the book in many meaningful ways, and I remain amazed after all these years at her
great intellect and breadth, which have been of such immense benefit to me in the form
of lively discussions of theory, method, and substance. She is my role model in every
way. And a special note of recognition to my daughter, Sarita, who inspires me and
serves as my role model every bit as much as her mother.

James Jaccard
Brief Contents

PART I. BASIC CONCEPTS

1 · Introduction 3

2 · The Nature of Understanding 7

3 · Science as an Approach to Understanding 22

PART II. CORE PROCESSES

4 · Creativity and the Generation of Ideas 51

5 · Focusing Concepts 95

6 · Thought Experiments for Variable Relationships 111

PART III. FRAMEWORKS FOR THEORY CONSTRUCTION

7 · Causal Models 151

8 · Mathematical Modeling 196

9 · Simulation as a Theory Development Method 248

10 · Emergent Theory: 267

Qualitative/Mixed‑Methods Approaches

xv
xvi Brief Contents

11 · Emergent Theory: Quantitative Approaches 307

12 · Historically Influential Systems of Thought 340

PART IV. THEORY AT THE LEVEL OF MEASUREMENT

13 · Theory and Measurement: General Frameworks 375

14 · Theory and Measurement: 402

Types of Measurement Strategies

PART V. CONCLUDING ISSUES

15 · Theory Revision 435

16 · Reading and Writing about Theories 461

17 · Epilogue 485

References 491

Author Index 511

Subject Index 518

About the Authors 522

Extended Contents

PART I. BASIC CONCEPTS

1 · Introduction 3
Organization of the Book / 4
Theories and Settings / 5

2 · The Nature of Understanding 7

The Nature of Reality / 8
Concepts: The Building Blocks of Understanding / 11
Conceptual Systems: The Bases for Deeper Understanding / 15
BOX 2.1. Concepts, Cultures, and Values / 15
Communication / 17
Summary and Concluding Comments / 19
Suggested Readings / 20
Key Terms / 21
Exercises / 21

3 · Science as an Approach to Understanding 22

Approaches to Understanding / 22
BOX 3.1. The Fringes of Science / 25
The Essentials of Scientific Endeavor / 26
Science and Objectivity / 27
The Process of Theory Construction / 28
Characteristics of a Good Theory / 32
What Is a Theoretical Contribution? / 33
Summary and Concluding Comments / 45
Suggested Readings / 46
Key Terms / 47
Exercises / 47

xvii
xviii Extended Contents

PART II. CORE PROCESSES

4 · Creativity and the Generation of Ideas 51
One Small Step for Science / 52
Creativity / 52
Choosing What to Theorize About / 56
Literature Reviews / 58
Heuristics for Generating Ideas / 59
BOX 4.1. The Nacirema / 60
Scientists on Scientific Theorizing / 81
Conceptual Logic Models / 84
Summary and Concluding Comments / 89
Suggested Readings / 90
Key Terms / 91
Exercises / 91
APPENDIX 4.1. Examples of Weak Argumentation / 93

5 · Focusing Concepts 95
The Process of Instantiation / 95
Shared Meaning, Surplus Meaning, and Nomological Networks / 98
Practical Strategies for Specifying Conceptual Definitions / 99
BOX 5.1. Etic and Emic Constructs / 100
Multidimensional Constructs / 103
Creating Constructs / 104
An Example of Specifying Conceptual Definitions / 105
Operationism / 107
Summary and Concluding Comments / 108
Suggested Readings / 109
Key Terms / 109
Exercises / 110

6 · Thought Experiments 111

for Variable Relationships
Thought Experiments for Relationships in Grounded
and Emergent Theory / 113
Describing Relationships with Different Types of Variables / 114
Thought Experiments for Relationships between Nominal Variables / 115
Thought Experiments for Relationships between Quantitative Variables / 118
Thought Experiments for Relationships between Nominal
and Quantitative Variables / 129
BOX 6.1. Simpson’s Paradox / 130
Thought Experiments for Moderated Relationships / 136
Broader Uses of Hypothetical Factorial Designs in Thought Experiments / 143
Summary and Concluding Comments / 146
Suggested Readings / 146
Key Terms / 147
Exercises / 147
 Extended Contents xix

PART III. FRAMEWORKS FOR THEORY CONSTRUCTION

7 · Causal Models 151
Two Types of Relationships: Predictive and Causal / 152
Causality and Grounded/Emergent Theory / 155
Types of Causal Relationships / 155
Constructing Theories with Causal Relationships / 158
Identifying Outcome Variables / 159
Identifying Direct Causes / 159
Indirect Causal Relationships / 160
Moderated Causal Relationships / 165
Reciprocal or Bidirectional Causality / 169
Spurious Relationships / 173
Unanalyzed Relationships / 175
Expanding the Theory Yet Further / 176
BOX 7.1. Finding Sources for a Literature Review / 184
The Binder Method / 187
Common Mistakes during Causal Theory Construction / 189
Perspectives on the Construction of Causal Theories / 189
Summary and Concluding Comments / 192
Suggested Readings / 192
Key Terms / 193
Exercises / 194

8 · Mathematical Modeling 196

Types of Variables: Nominal, Discrete, and Continuous / 197
Axioms and Theorems / 198
Functions / 198
Linear Functions / 199
Deterministic versus Stochastic Models / 203
Model Parameters / 204
Rates and Change: Derivatives and Differentiation / 205
Describing Accumulation: Integrals and Integration / 208
Just‑Identified, Overidentified, and Underidentified Models / 209
Metrics / 210
Types of Nonlinearity / 210
BOX 8.1. Reading Mathematical Models / 211
Functions for Nominal Variables / 222
Advanced Topics: Manipulating and Combining Functions / 224
Multiple Variable Functions / 227
Phases in Building a Mathematical Model / 228
An Example Using Performance, Ability, and Motivation / 228
An Example Using Attitude Change / 232
Chaos Theory / 235
Catastrophe Theory / 237
Additional Examples of Mathematical Models in the Social Sciences / 238
Emergent Theory Construction and Mathematical Models / 238
Summary and Concluding Comments / 239
xx Extended Contents

Suggested Readings / 240

Key Terms / 241
Exercises / 242
APPENDIX 8.1. SPSS Code for Exploring Distribution Properties / 243
APPENDIX 8.2. Additional Modeling Issues for the Performance,
Ability, and Motivation Example / 245

9 · Simulation as a Theory Development Method 248

Defining Simulations / 249
The Uses of Research Simulations / 250
The Difference between Simulations and Laboratory Experiments / 250
Basic Simulation Varieties / 251
The Analysis of Criterion Systems as a Basis for Theory Construction / 253
Simulations and Virtual Experiments / 260
Agent‑Based Modeling / 260
BOX 9.1. Agent-Based Modeling of Segregation / 262
Resources for Conducting Simulations / 263
Summary and Concluding Comments / 263
Suggested Readings / 264
Key Terms / 265
Exercises / 265

10 · Emergent Theory: 267

Qualitative/Mixed‑Methods Approaches
Grounded and Emergent Theory: An Overview / 268
Qualitative Research and Ways of Knowing / 269
Framing the Problem / 270
The Role of Past Literature / 271
Collecting Qualitative Data / 271
Field Notes and Memo Writing / 277
BOX 10.1. Anthropology and the Ethnographic Tradition / 278
Theoretical Sampling / 279
Analyzing and Coding Data / 279
Constructing Theory in Qualitative Research / 285
Mixed‑Methods Research / 291
Products of Qualitative and Mixed‑Methods Research / 293
Summary and Concluding Comments / 294
Suggested Readings / 296
Key Terms / 297
Exercises / 298
APPENDIX 10.1. The Limits of Information Processing / 300

11 · Emergent Theory: Quantitative Approaches 307

Exploratory Analyses of Direct Effects / 309
BOX 11.1. Robust Statistics / 318
Exploratory Analyses of Moderated Relationships / 320
Cluster Analysis / 325
Factor Analysis / 329
 Extended Contents xxi

Big Data, Data Mining, and Machine Learning / 331

Chance Results / 335
Summary and Concluding Comments / 336
Suggested Readings / 336
Key Terms / 337
Exercises / 338

12 · Historically Influential Systems of Thought 340

Grand Theories / 341
Frameworks Using Metaphors / 352
Frameworks Emphasizing Stability and Change / 357
Psychological Frameworks / 359
BOX 12.1. Collaboration / 360
Frameworks Inspired by Methodology / 364
Summary and Concluding Comments / 369
Suggested Readings / 369
Key Terms / 371
Exercises / 372

PART IV. THEORY AT THE LEVEL OF MEASUREMENT

13 · Theory and Measurement: General Frameworks 375
Defining Measurement / 376
Conceptual Definitions and Measurement / 377
Classic Test Theory: The Basics / 379
The Facets of Measurement and Theory Construction / 382
Measurement Theory Construction / 383
Emergent Theory Building in Measurement: Cognitive Testing,
Expert Interviews, and Pilot Research / 390
BOX 13.1. Measuring Sex and Gender / 391
Using Extant Measures / 393
Other Criteria for Measure Evaluation / 395
Classical Test Theory and Qualitative Research / 396
Writing Reports to Justify Measurement Choices / 396
Empirical Tests of Measurement Theory / 397
Summary and Concluding Comments / 397
Suggested Readings / 399
Key Terms / 400
Exercises / 400

14 · Theory and Measurement: 402

Types of Measurement Strategies
Constructing a Theory of Self‑Reports / 403
BOX 14.1. Quantifying Love / 413
Constructing a Theory of Observer Reports / 423
Constructing a Theory of “Objective” Measures / 425
The Importance of a Theory Construction Mindset to Measurement / 426
Measurement and Qualitative Research / 426
xxii Extended Contents

Summary and Concluding Comments / 428

Suggested Readings / 429
Key Terms / 430
Exercises / 431

PART V. CONCLUDING ISSUES

15 · Theory Revision 435
Disconfirmation and Theory Revision / 436
Boundary Conditions and Theory Revision / 438
Replication and Theory Revision / 440
Disconfirmation and Qualitative Research / 441
BOX 15.1. Strong Inference / 442
Bayesian Perspectives in Theory Revision / 443
A Look to the Future: Computer Automation and Theory Revision / 451
Theoretical Desiderata for Revised Theories / 456
Paradigm Shifts / 456
Summary and Concluding Comments / 457
Suggested Readings / 458
Key Terms / 458
Exercises / 459

16 · Reading and Writing about Theories 461

Reading about Theories / 461
BOX 16.1. PowerPoint Presentations of Theories / 466
Writing about Theories / 468
Grant Proposals, Technical Reports, and Presentations / 473
Summary and Concluding Comments / 474
Suggested Readings / 475
Key Terms / 475
Exercises / 476
APPENDIX 16.1. Inferring Theoretical Relationships
from the Choice of Statistical Tests / 477

17 · Epilogue 485
A Program of Self‑Study / 488
Concluding Comments / 490

References 491
Author Index 511
Subject Index 518
About the Authors 522

The companion website (www.theory-­construction.com) includes PowerPoint slides

of all of the book’s figures, methodological primers and video demonstrations,
supplemental exercises, and other resources.
Part I

BASIC CONCEPTS
1

Introduction

Few people dispute the central role of theory in the social sciences. Scientists formulate
theories, test theories, accept theories, reject theories, modify theories, and use theories
as guides to understanding and predicting events in the world about them. A great deal
has been written about the nature and role of theory in the social sciences. These writ-
ings have spanned numerous disciplines, including anthropology, economics, history,
philosophy, political science, psychology, sociology, and social work, to name but a few.
This literature has described, among other things, broad frameworks for classifying
types of theories, the evolution of theories over time, the lives and scientific strategies
of great scientific theorists, and general issues in the philosophy of science. Although
this literature is insightful, much less has been written to provide social scientists with
practical guidelines for constructing theories as they go about the business of doing
their science. Most students are intimidated by the prospect of constructing their own
theories about a phenomenon. Theory construction is viewed as a mysterious process
that somehow “happens” and is beyond the scope and training of a young scientist try-
ing to find his or her way in the field. Whereas most graduate programs in the social
sciences require multiple courses in research methodology so that students will become
equipped with the tools to test theories empirically, the same cannot be said for theory
construction. In contrast to focusing on methods for testing theory, the current work
focuses on methods for generating theory.
The fundamental objective of this book is to provide students and young scientists
with tools to assist them in the practical process of constructing theories. It does so via
describing in some detail the strategies, heuristics, and approaches to thinking about
problems that we have found to be useful over the more than 70 collective years that
we have been doing social science research. This book is not an academic discussion of
the literature on theory construction or the philosophy of science. We do not delve too
deeply into the vast literature on these topics. Rather, we take a more practical journey
through the cognitive heuristics, tricks of the trade, and ways of thinking that we have
found to be useful in developing theories.

3
Another Random Scribd Document
with Unrelated Content
2, pp. 488-489. 2 Dall in Cont. N. A. Eth., vol. i, pp. 93-106. ^ Bull.
40, B. A. E., pt. I, pp. 971-972. i
 SWANTON- dixon! PRIMITIVE AMERICAN HISTORY 2>97
Traditional and carl\- liistorical references as well as similarity in
language all ixjint to a separation of the last mentioned body from
the Skidi Pawnee at a comparatively recent period. Of the Pawnee
tribes proper the Skidi were to the north of the others and seem to
have considered themselves original inhabitants of the countr>-
occupied 1)>- them when first discovered. According to Mr James
Murie two of the remaining tribes placed their original homes in the
east, one as far as the Ohio, while the last claimed to have come
from the southwest. The Wichita are merely the largest and most
representative of a group of seven or eight allied peoples most of
whom have been absorbed by them. When first encountered by
whites they were camping along Arkansas river and its branches.^
Late in the eighteenth or early in the nineteenth century, however,
they were pressed out of this country by northern and eastern tribes
and moved southwest, first to the North Canadian, later to the
Wichita mountains.^ There is no tradition pointing to any region
outside of this area. The Kichai were formerly on the upper waters of
Red river whence they were gradually forced down upon the Trinity.
No Kichai migration legend has come to our attention. The Caddo
proper also seem to have partaken of the comparatively immobile
character of the tribes of this stock. They were found by the De Soto
expedition, in the region later associated with them, and there is no
legend pointing to a place of origin or habitation anywhere beyond.
Sibley cites a tradition to the effect that the Kadohadatcho, the
leading eastern Caddo tribe, had formerly lived at the Cross Timbers,
375 miles above their later seats,^ but this does not indicate any
general movement on the part of all of the tribes. An origin myth
collected by one of the writers from a Natchitoches Indian teikes us
back to the neighborhood of Lake Sodo. • Handbook of Am. Indians,
article Quivira. La Harpe in Margry, Decouverles, vol. VI, p. 289. 2
Gatschet in Am. Antiq., Sept. 1891, pp. 249-252. * Annals of Cong..
9th Cong., 2d s^ss., 1085.
 -,g8 AMERICAN ANTHROPOLOGIST [n. s., i6, 1914 X. —
Indians of Southern Texas South of the Caddoan peoples were a vast
number of Indian tribes now classified into three linguistic stocks
called Tonkawan, Karankawan, and Coahuiltecan, but there are
reasons for believing that more complete linguistic data (which
unfortunately it will be difficult to obtain from any but the first
mentioned) would show these to be related. And it is also probable
that they would be found to have a connection with the ancient
inhabitants of the northern and central parts of the state of
Tamaulipas, Mexico. Further than this we have practically no
information, no migration traditions having been preserved and little
information of any kind regarding them having been recorded. XI. —
The Kiowa The Kiowa, constituting the Kiowan linguistic stock, are
associated in history with the southern plains, but about 1780 they
were in the Black hills and their own traditions as recorded by
Mooney carry them back to the head waters of the Missouri in
western Montana. Mooney believes that their affiliation is rather with
the tribes west of the Rocky mountains than with those on the
eastern side, and recent investigations would seem to confirm this
view.^ XII. — Indians of the Athapascan Stock We now turn to the
great western division of stocks referred to at the beginning of this
paper. In point of territory covered, the Athapascan family equals, if
indeed it does not outrank, the Algonquian, which is usually
considered the largest of all the stocks in North America.
Geographically the Athapascans fall into three separate groups,
Northern, Pacific, and Southern. The first, and by far the largest of
these, comprises the various tribes sometimes known collectively as
Tinneh or Dene. In one immense continuous area they spread over
the whole of the interior of Alaska, northern British Columbia, and
the Mackenzie basin, extending over about 65° of longitude and 1
Mooney in 17th Ann. Rep. B. A. E., pp. iSi-iSS- See J. P. Harrington in
Am. Anth., XII. 119-123 
 swanton-dixon] primitive AMERICAN HISTORY 399 nearly
20° of latitude. Among the more important of their many tribes were
the Dog-ribs, Yellow-knives, Chipewyans, the various Kutchin
divisions, the Xahane, Carrier, and Chilcotin. A small isolated tribe,
the Sarsi, lived with the Algonquian Blackfoot in southeastern Alberta
and northern Montana. The Pacific group includes a small isolated
band in southern British Columbia, together with others in western
Washington, and a series of small tribes stretching in a nearly
continuous strip along the Oregon and California coasts between
Umpqua and Eel rivers. The southern division, of which the most
important members were the Navaho and Apache, occupied a large
area in eastern Arizona, western and southern New Mexico, and
southwestern Texas extending southward some distance into the
Mexican states of Chihuahua and Coahuila. A small isolated group of
Athapascan people, the Kiowa Apache, were with the Kiowa in the
southern Plains. The historical problems presented by the
Athapascan stock are among the most difficult as well as most
interesting in the northern continent, and there is much difference of
opinion not only in regard to the movements of the various
individual tribes and branches, but also concerning the relations of
these branches within the stock. For the northern branch, migration
traditions have been recorded chiefly from the tribes of the
Mackenzie basinThese were first given by Mackenzie himself ^ and
have since been secured by others, notably by Petitot.- Most of these
accounts seem to be in accord in placing their earlier home far to the
west, either across the sea or on the other side of a long lake full of
islands. From this western land they were driven by the cruelty and
fierceness of their neighbors, and after long travel and many
difficulties came into their historical habitat. Some versions of the
tradition make this western home a sort of terrestrial paradise, and it
is uncertain how far the accounts are to be taken as purely mythical.
Little or no information has been gathered from the Alaskan tribes as
yet, and until more abundant material is at hand, it is premature to
try to draw conclusions. The most that may be said is that '
Mackenzie, Voyages, etc., p. cxviii. 2 Petitot, Monographie des Dene-
Dindjie.
 400 AMERICAN ANTHROPOLOGIST [n. s., i6. 1914 attempts
to derive the northern Athapascans from Asia on the basis of these
traditions are absurd. The only realh' definite indication of migration
in this northern group is in the southward movement of the Sarsi,
who separated from the main body to the north, and allied
themselves with the Blackfoot. A similar origin seems to be indicated
for the small tribe formerh- living in the Nicola valley in southern
British Columbia. The scattered tribes or bands forming the Pacific
group seem to possess no trace of any traditions of migration, and
all, without exception so far as is known, locate the creation of their
first ancestors within the territory where the bands were living at the
time of first European contact. Their general distribution, however, is
such as to indicate a movement parallel to the coast and
presumably, in conformity with other tribes in this region, from north
to south. From the completeness of their adaptation to the
environment it would seem that the original immigration into this
coastal area must have taken place at an early period. The two great
tribes which together comprise the larger portion of the southern
group present an interesting problem. Two contrasted points of view
are held. Hodge, ^ relying on the statements of early Spanish
writers and explorers as well as native traditions, believes that the
Apache moved westward from eastern New Mexico and had not
reached Arizona until after the middle of the i6th century. On this
theory they would be thus comparatively recent comers in the
Southwest, where they have, with the usual readiness of the tribes
of Athapascan stock, adapted themselves rapidly to their new
environment, and borrowed many elements of their culture from the
sedentary Pueblo tribes with which they came in contact and
portions of which they completely absorbed. The Navaho on this
theory are believed to have appeared originally about the end of the
15th century in northern New Mexico. At first an insignificant tribe,
they grew gradually, in part by absorption of other elements derived
from the Rio Grande pueblos, the Zufii, the Ute, and the Yuman
stock, and in part by incorporation of portions of the affiliated
Apache, and in this way extended their 1 Hodge, The Early Navaho
and Apache, Am. Anlhr., 1895, viii, pp. 223-240.
 SWANTON DIXON] PRIMITIVE AMERICAN HISTORY 40I
territory westward far into Arizona. Goddard ' on the other hand,
relying more on cultural and linguistic considerations, believes that
the evidence brought forward by Hodge is inconclusive, and that the
Apache and Navaho are on the contrary old residents of the
Southwest, having become completely assimilated to the
environment in a way imijossible if they were recent comers. The
migration and origin legends regarded by Hodge as in large part
really historical are thus considered to be almost wholly mythical and
to have little or no value as indicating tribal movements. The final
solution of this jiroblcm must await fuller archeological evidence. For
the small isolated tribe of the Kiowa Apache — whose afifiliations
seem clearly with the northern group — ^we have distinct traditions
of their meeting with the Kiowa at the time when these were still in
Montana, and of their accompanying them in their southward
movements in the Plains. The larger problem of the movement of
the Athapascan stock as a whole has usually been answered by
assuming a southerly drift by which portions, breaking away from
the parent body in the north, have wandered southward through the
Plains as far as New Mexico and Arizona, the Sarsi and Kiowa
Apache being laggards or remnants left behind. The Pacific group
were thought to be either jjortions of these who passed west across
the Rockies, perhaps down the Columbia, and then from its mouth
down the coast as far as California, or else a separate migration
from the westerly portion of the northern parent stock passing
directly south along the Pacific shores, and of which the Washington
and southern British Columbian fragments represented the laggards
or latest comers. This view has been opposed by Goddard - who
believes that the exact contrary is not improbable, and suggests that
a further possibility is that the stock formerly had a continuous
distribution but has been disrupted by the intrusion of other peoples.
Until, however, more conclusive proof in favor of a northward
movement or of a disruption by force is brought forward, the theory
of a southerh- drift seems best to fit the facts. * Goddard, XVUi
Congress of Americanists, 1, pp. 337-359 2 Ibid.
 ^02 AMERICAN ANTHROPOLOGIST [n. s., i6. 1914 XIII. —
Indians of the North Pacific Coast We nuu- tlixidc the Indians of the
north Pacific coast rout;iily into two sections, a northern composed
of the Chimmesyan, Skittagetan, and Koluschan stocks, and a
southern, mainly represented by the Wakashan and part of the
Salishan peoples. Among the former the Chimmes>'ans stand
entirely apart, linguistically and to a certain extent culturally. They
consist of three tribes, the Tsimshian on Skeena river, the Niska on
Nass river, and the Kitksan on the headwaters of both these streams.
Although typically a coast people their traditions all point to an
inland origin, at least as far back from the coast as the present
territory of the Kitksan. The Skittagetan stock, embracing the people
more often known as Haida, was located on the Queen Charlotte
islands, British Columbia, and the southern end of Prince of Wales
island, Alaska. The traditions, both of the Haida themselves and the
other Alaskan Indians, show that those Haida now on Prince of
Wales island emigrated to that region some time in the early part of
the eighteenth century. 1 The traditions of the Queen Charlotte
Islands Haida carry us to the eastern shore of the islands,
particularly to the northeastern point and to the southern end.^ The
Koluschan stock, embracing the Indians usually known as Tlingit,
extended over all the coast and islands of the panhandle of Alaska,
with the exception just indicated, and beyond as far as the mouth of
Copper river. The traditions of the greater number of their clans
point to an origin on the Nass river to the south, but that of the
K!ackeqoan brings them from among the Athapascan tribes on
Copper river, that of the Nanyaayi points to an origin inland from
Taku inlet, and that of the Qatcadi to the interior along the upper
Skeena.^ On the other hand several Tlingit clans are now
represented among the Tahltan of the upper Skeena by later
settlement or intermarriage from the coast, ^ and the Tagish of
Chilkat pass are said to be a Tlingit offshoot.-"^ This last statement,
however, is probably an 1 Dawson in Rep. Geol. Survey Can., for
1879, p. 104B. Swanton in Metn. Am. Mus. Nat. Hist., viii, pp. 88-90.
2 Swanton, ibid., p. 72 et seq. 3 Swanton in 26th Ann. Rep. B. A. E.,
p. 410; also cf. p. 411■> Emmons in Anth. Pub. Univ. of Pa., vol. IV,
no. i, pp. 11-21. 5 Dawson in Rep. Geol. Surv. Can., 192B. 1887.
 SWANTON-DIXON] PRIMITIVE AMERICAN HISTORY 403
error. Within comparatively late historic times the Tlingit have moved
farther west toward Copper river, and have modified an Eskimo tribe
on Kayak island, the Ugalakmiut, to such an extent that these are
now indistinguishable from the Tlingit proper, having adopted their
language as well as their customs. ^ The Tlingit and Haida
languages furnish still further evidence of an inland origin, the
resemblance between at least Tlingit and Athapascan being very
marked. The Wakashans consist of two branches, the Kwakiutl of
Queen Charlotte sound and the coast northward to Kitamat, and the
Nootka of the west coast of Vancouver and the extreme
northwestern point of Washington. Many of these tribes are divided
into family groups which trace their origin from an ancestor who
descended from the sky and settled at such and such a place. As
village sites are usually to be found at the places indicated it is
probable that they were in fact formerly occupied by the people in
question. Nevertheless these sites are all in the same region and do
not indicate any movement en masse from elsewhere.The Salishan
tribes may be divided roughly into the coast Salish and the interior
Salish. The former were on Georgian straits, the Straits of Fuca,
Puget sound, and on the outer coasts of Washington and Oregon —
with the exception of the Columbia river entrance, and the
northwestern corner of Washington stateas far south as Siletz river.
Still farther north, on North and South Bentinck arm, Dean inlet, and
Bellacoola river, was a detached body known as the Bellacoola.
These seem to have migrated from the coast Salish farther south,
but along the heads of the deep inlets instead of by the outer coast.
The interior Salish occupied a large part of the lower Frazer valley,
including the valle\- of the Thompson, the upper valley of the
Columbia, and as far east as the headwaters of the Missouri. While
no memory appears to have been preserved of movements among
these people in great bodies, there is reason to believe that the
coast Salish originally pressed down from the interior. At least Boas
is able 1 Petroff in Tenth Census, vol. viii, p. 146. 2 Boas in Rep. U.
S. Nat. Mus., for 1895. pp. 328-334 
 404 AMERICAN ANTHROPOLOGIST [n. s., i6, 1914 to say
that "both linguistic and archaeological indications suggest that the
Salish tribes which now inhabit the coast of the Gulf of Georgia
separated from the Salish tribes of the interior at a time when both
had the simple form of culture that seems to be characteristic of the
whole plateau area and of the Mackenzie basin." ^ The Chimakuan
stock consists, or rather consisted, of but two tribes, the Chimakum
about Port Townsend, Washington, and the Quileute on the
northwestern coast of the same state. It is believed that a closer
study of the Chimakuan language may show some connection with
Salish. XIV. — The Kutenai The Kitunahan stock consisted of the
Kutenai tribe only. Its historic seat was in southeastern British
Columbia along the west flanks of the Rocky mountains, extending
also slightly into the present United States. Chamberlain says
regarding the origin of these people: "Their traditions suggest that
they are comparatively modern intruders into this area from some
quarter to the east of the Rockies, possibly around the headwaters
of the Saskatchewan." ^ Their language shows some points of
resemblance with those of the Shoshonean group. XV. — The
Shahaptians and the Indians of Western Oregon The Shahaptian
area included a considerable territory in the vicinity of the Columbia
and Snake rivers, in southwestern Idaho, southeastern Washington,
and northeastern Oregon. The best. known of the several tribes
composing the stock was the Nez Perce. Very little information is
available in regard to the early history of these tribes, which were
first met by Lewis and Clark at the beginning of the last century. The
Nez Perce themselves seem to have been long in their historic
habitat; on the other hand the Klikitat appear to have begun a
movement westward across the Cascades not long before European
contact, and to have thus paralleled north of the Columbia the
movements of the Molala south of it. 1 Ann. Arch. Rep. for 1905,
App. to the Rep. of the Minister of Education, Ontario, p. 225. 2
Chamberlain in Ann. Arch. Rep., op. cit., p. 178.
 swanton-dixon] primitive AMERICAN HISTORY 405 A
number of small, apparenth- independent linguistic stocks occupied
the western portion of Oregon at the time \\lKn it first became
known to Europeans. These were the Chinookan along both banks of
the Columbia from the Dalles to the sea; the Kalapooian in the
Willamette valley; the Kusan about Coos bay; the Siuslauan and
Yakonan just north of these along the coast; the Takelman isolated
among Athapascan peoples on the middle Umpqua; the Waiilatpuan
in two separate areas, one along the western slope of the Cascades
south of the Columbia, and one southeast of the bend of the
Columbia at Wallula; and lastly the Lutuamian, who occupied the
southern Cascades, mainly on their eastern slope, and the basins of
the Klamath lakes. For the majority of these, no traditional or other
evidence of migration is available. Exceptions are in the case of the
Molala who are said by the Cayuse (the eastern branch) to have
separated from them, and to have crossed the Cascades toward the
west to their historic sites. As the two dialects are quite distinct, this
separation must have occurred at an early time. The Klikitat and
some other Shahaptian tribes also seem to have been pushing north
and west.^ For the Kalapooians there is some evidence of a
southward movement of slight extent, toward Umpqua valley. XVI.
— Indians of California The Californian area presents a somewhat
troublesome problem. Powell divided the languages of the state into
twenty-two separate stocks, with the result that this region appeared
to be linguistically one of the most complex in the world. Recent
investigations however by Kroeber^ and one of the authors and also
by Sapir^ make it very probable that the many stocks of Powell may
be reduced to nine or ten, of which three (Shoshoncan, Athapascan,
and possibly Algonquian) are mainly extra-Californian families. Of the
newly determined families, the largest is the Penutian, occupying a
continuous area which ma>- be roughly described as > Lewis, Mem.
Am. Anthr. Asso., vol. i, pt. 2, pp. 195-196. Gibbs in Coiil. to N. A.
Eth., vol. I, p. 224. 2 Dixon and Kroeber. Amer. Anlhr. (n. s.), xv. pp.
647-655. 'Sapir, ditto, pp. 617-646.
 406 AMERICAN ANTHROPOLOGIST [n. s.. i6. 191 4
incliuling- tlic whole of the (ireat Valley together with the coastal
region south of San Francisco to beyond Monterey. This includes the
former Wintun, Maidu, Miwok, Costanoan, and Yokuts stocks. No
definite traditions of migration have been found among any of the
members of the Penutian family, but on linguistic grounds there
would seem to be some e\idence of a former continuity of the Maidu
and Yokuts groups, now separated by the intervening Miwok; and in
general of a spreading outward from the central portion of the state
along the courses of the Sacramento and San Joaquin rivers. The
second large Californian stock is the Hokan, whose territory is much
broken up. In the north it comprises the region occupied by the
Shastan, Chimarikan, and probably the Karok and Yanan groups as
well. Separated from these and farther south are the Pomo, along
the coast and in the Coast Ranges north of San Francisco; the now
extinct Esselen on the coast south of Monterey; and the Yuman
group of the extreme south of the state and in western Arizona. As
in the case of the Penutian stock, practically no traditional evidence
is available indicating any migratory movements except the slight
indications shown by the Yuman branch. The area occupied by
Yuman tribes comprised southw^estern Arizona, the extreme
southern portion of California, and the northern portion of the
peninsula of Lower California. As in the case of most tribes west of
the Rockies, there is little traditional evidence of migration. In one or
two cases, however, there are some facts which may be significant.
Thus the Havasupai now* living in Cataract canyon (a tributary of
the Colorado just west of the Grand canyon) have traditions of
having lived formerly farther to the south, along the Little Colorado
and upper Verde rivers. The Yavapai on the other hand, would seem
to have moved from a position along the Colorado near the mouth of
Bill Williams fork, eastward toward central Arizona. A somewhat
similar eastward movement also occurred in the case of the
Maricopa who moved during the 19th century from a position near
the mouth of the Gila to one near its middle course. Except for the
Havasupai, who acquired not a little of the characteristic culture
features of
 swanton-dixon] primitive AMERICAN HISTORY 407 the
Pueblo tribes, the general type of Yuman culture is reminiscent of
California, and would suggest an earlier home in that direction. The
Shastan group shows some indications of a southerly movement,
and general considerations — cultural, linguistic, and geographic—
make the supposition of a similar tendency for the whole stock
probable. How far the intrusion of the Athapascans has been
responsible for this it is as yet impossible to say; the possibility of
disruption due to this cause and to the expansion of the Penutian
stock must certainly be considered. It seems probable, however, that
any such movements, both in this case and in that of the Penutian
stock, must have taken place at a very early period. For the other
Californian stocks, there is little evidence at hand. The Yuki, who are
in three separate divisions, two north of and one south of the Pomo,
show evidence of disruption by the intrusive Athapascans, and of an
older separation by which the southern or Wappo group were
divided from the parent stock. The Washo in the region about Lake
Tahoe on the eastern border of the state show no indications of
movement in any direction. For the Salinan and Chumash stocks of
the southern coast also there is no traditional or other evidence
which would show tribal movements, and it is probable that they
have been for a very long period in occupancy of the region in which
they were found by the earliest European explorers. XVII. — Indians
of the Shoshonean Stock The Shoshonean tribes stand at present in
a somewhat uncertain position as regards their linguistic
independence. Since the middle of the last century ^ a feeling has
been growing that the Shoshonean languages should be grouped
with the Piman and Nahuan to form a larger stock or family, called
by Brinton - the Uto-Aztecan. Leaving this question aside for the
moment, however, the history of the Shoshonean branch can be
briefly summarized. The area covered by tribes of this group at the
time of their * Buschmann, Spur en der aztekischen Sprache, Berlin,
1859. ^ Brinton, American Race, p. 118 sq. AM. ANTH., N. S., l6 —
27
 4o8 AMERICAN ANTHROPOLOGIST [n. s., i6. 1914 earliest
contact with Europeans was, with two exceptions, a continuous one.
The mass of the people lived almost wholly within the region
generally known as the Great Plateau, and comprised southeastern
Oregon, southern Idaho, southwestern Montana, western Wyoming
and Colorado, the whole of Utah and Nevada, together with most of
California south of the Tehachapi and a narrow strip along its
eastern border. The two outlying tribes were the Hopi, whose
villages lay in northern Arizona, and the Comanche, who ranged over
the southern Plains. On a linguistic basis 1 the Shoshonean tribes
may be divided into four very unequal subdivisions: the Pueblo
(comprising the Hopi only); the Plateau (the most important tribes
being the Ute, Shoshoni, Comanche, and Paiute); the Kern River;
and the Southern California (including the Serrano, Gabrieleno,
Luisefio, Cahuilla, etc.). Little has been recorded for any of these
tribes, except the Hopi and Comanche, in the way of migration
traditions. The Hopi were of complex origin, and will be considered
along with the other Pueblo Indians. The Comanche are linguistically
closely related to the Shoshoni of Idaho and Wyoming, and there is
traditional evidence- of their being residents of that section early in
the 1 8th century, and that they were driven by other tribes from
this northern home southward along the western edge of the Plains.
At this same period, probably, the Shoshoni were forced west across
the Rockies to their hstorical site. Brinton ^ and others have held
that this latter movement indicated a former residence of the whole
stock in the region between the mountains and the Great Lakes; and
Powers ^ supposed the southern California tribes to be recent
intruders there from the eastward. There seems, however, to be little
ground for either of these assumptions, and the evidence, both
linguistic and cultural, would appear to show that the tribes
composing the Shoshonean group have been in occupancy of the
Great Plateau and of southern California for a very long time. 1
Kroeber, Univ. Col. Pub. Amer. Arch, and Eth., iv, p. 97 et seq. 2
Clark, Indian Sign Language, p. 118. 3 Op. cit., p. 121. 4 Powers,
Tribes of California, p. 369.
 SWANTON-DixoN] PRIMITIVE AMERICAN HISTORY 409
XVIII. — Indians of the Piman Stock The Piman family holds still, like
the Shoshonean, a somewhat uncertain position in regard to its
linguistic independence, and it is probable ' that with the
Shoshonean and Nahuan it forms merely a branch of the larger Utb-
Aztecan stock. The larger part of the territory occupied b>- tliis
group lies in northwestern Mexico, in the states of Sonora,
Chihuahua, Sinaloa, and Durango, with extensions still farther south;
of the tribes in the United States the Pima and Papago are the most
important, and occupied in the 1 8th century a considerable area in
southern Arizona. The origin tradition of the Pima- refers to the Salt
River valley as the region where the tribe had its beginning, and
states that their ancestors moved thence southward to the Gila;
much later, under the attack of enemies from the east, a portion
moved into Mexico while others went northward to join the Zuiii and
Hopi.-^ Other traditions refer to an earlier eastern home.^ That the
Pima had been long settled in the southern portion of Arizona seems
indicated by the abundant ruins throughout the area, the majority of
which, including the famous Casa Grande, are attributed to their
ancestors.^ The fact that linguistically the Piman languages stand
closer to the Shoshonean than they do to the Nahuan dialects ^ and
that geographically they are intermediate between these two
branches of the Uto-Aztecan family, may perhaps be taken as
indicating a general southerly drift for the entire great group. More
definite knowledge of the culture and archeology of northwestern
Mexico is, however, necessary before any certain conclusions can be
reached. XIX. — The Pueblo Indians There is very little information
available regarding the migration traditions of the Pueblo Indians
outside of the Hopi and the Zuni. All that we can make out is a
widespread belief that the ' Kroeber, op. cit., p. 164. 2 Russell in 26lh
Ann. Rep. B. A. E., pp. 206-23C. ' Fewkes in 28lh Ann. Rep. B. A. E.,
pp. 153-160. * Russell, op. cit., p. 26. ' Fewkes, loc. cit. • Kroeber,
op. cic, p. 163.
 4IO AMERICAN ANTHROPOLOGIST [n. s., i6, 1914 people
had come up from the underworld at some point in the north.
According to dishing the Zuiii were composed of two elements, an
earlier element, the traditional origin of \\lu(li was identical with that
given above, and a later element from the west or southwest.^
According to Dr Few^kes the Hopi were formed by three prehistoric
immigrations, the first of which, consisting of the Honau or Bear
people and Kokop or Firewood people, he believes to have come
from the Rio Grande region, tradition specifying Jemez. Secondly
came the Snake people from the San Juan region in the north, who
settled first on the Little Colorado west of Walpi, and finally came to
Tusayan. The third and last consisted of what is now the Patki
people who came up from the Gila valley, and were perhaps of
Piman origin. They were very likely of the same stock as the
southern immigrants into Zuni. Within historic times, especially since
the rebellion of the Pueblos against the Spaniards in 1680, several
other movements ha\-e taken place. Thus the Asa, a Tewa people,
moved to Zuiii and from there again to Hopi, founding the Pueblo of
Sichomovi, called "the Zuni town." About 1710 came the Hano
people, also of Tewa stock, and founded the pueblo of that name
where the Tewa language is still preserved. Some Keres also came to
Hopi, but the bulk of them afterward left and founded Sandia. Over
and above these great migrations movements of small bodies of
persons frequently occurred, sometimes perhaps of two or three
people only, but this served to spread clans from one pueblo to
another and to increase the complexity throughout.^ XX.
COXCLUSION Let us now recapitulate briefly. From the data available
it appears that the origin of the tribes of several of our stocks may
be referred back to a swarming ground, usually of rather indefinite
size but none the less roughly indicated. That for the Muskhogeans,
including probably some of the smaller southern stocks, must be
placed in Louisiana, Arkansas, and perhaps the western parts of
Mississippi and Tennessee, although a few tribes seem to 1 Gushing
in 13th Ann. Rep. B. A. E., p. 342. 2 Fewkes in igih Ann. Rep. B. A.
E., pp. 573-634.
 swaxton-dixon] primitive AMERICAN HISTORY 4II have
come from the region of the Ohio. That for the Iroquoians would be
along the Ohio and perhaps farther west, and that of the Siouans on
the lower Ohio and the country to the north including part at least of
Wisconsin. The dispersion area for the Algonquians was farther
north about the Great Lakes and perhaps also the St Lawrence, and
that for the Eskimo about Hudson bay or between it and the
Mackenzie ri\-er. The Caddoan peoples seem to have been on the
southern plains from earliest times. On the north Pacific coast we
have indications that the flow of population has been from the
interior to the coast. This seems certain in the case of the Indians of
the Chimmesyan stock and some Tlingit subdivisions. Some Tlingit
clans, however, ha\'e moved from the neighborhood of the Nass
northward. Looking farther south we find evidence that the coast
Salish have moved from the inner side of the coast ranges, while a
small branch has subsequently passed northward to the west of it.
The Athapascan stock in all probability has moved southward,
sending one arm down the Pacific coast, and a larger body
presumably through the Plains which reached as far as northern
Mexico. Most of the stocks of the Great Plateau and of Oregon and
California show little evidence of movement, such indications as are
present, however, pointing toward the south as a rule. The Pueblo
Indians appear to have had a mixed origin, part of them coming
from the north, part from the south. In general there is to be noted
a striking contrast between the comparatively settled condition of
those tribes west of the Rocky mountains, and the numerous
movements, particular!},- in later times, of those to the east. While
we can hope for little more traditional evidence regarding the
migrations of our Indians the collection of further ethnological
material of all kinds is bound to cast a flood of light upon the whole
question of tribal movements. More exact information regarding
Indian languages will doubtless bring out new resemblances and
contrasts, some of which will in time be shown to have historic
value. Again, all of these tribes must be reclassified in accordance
with the data yielded by physical anthrojxjlogy as soon as those data
are sufficiently complete. We already know that this classi 
 412 AMERICAN ANTHROPOLOGIST [n. s., i6, 1914 fication
will show a ver>- different alignment of tribes, that in some cases
linguistic stocks will be cut to pieces and in other cases brought
together. This discordance, however, far from disturbing us, should
be welcomed as giving a different angle of approach which will
probably enrich rather than confuse our conception of aboriginal
American history. The study of cultural features properly so
considered will also yield certain valuable results, at least of
confirmatory value, but less is to be expected from this branch of
ethnology than from the two already considered. Culture, however,
as well as physical anthropology, has one great advantage over
language in that it can be enriched progressively by archeological
investigations long after the living peoples are extinct, and there will
come a time when the archeological method of approach will be the
only method remaining.
 AREAS OF AMERICAN CULTURE CHARACTERIZATION
TENTATIVELY OUTLINED AS AN AID IN THE STUDY OF THE
ANTIQUITIES^ By W. H. holmes Contents Introduction 413 The
North Atlantic Area 41? The Georgia-Florida Area 420 The Middle
and Lower Mississippi Valley Area 424 The Upper Mississippi and
Great Lakes Area 428 The Great Plains and Rocky Mountain Area 43°
The Arid Region 432 The California Area 435 The Columbia-Fraser
Area 43^ The Northwest Coast Area 440 The Arctic Shoreland Area
443 The Great Northern Interior Area 445 Introduction AS an initial
step in the description and interpretation of the antiquities of the
continent, the archeologist observes the tribes of today, their cultural
characteristics and environments, and acquaints himself with what is
known of them historically. He finds that their achievements are
greatly diversified and that certain forms and states of culture
characterize particular geographical areas and realizes that
environment has had a large share in determining the course of the
culture evolution. He examines the antiquities and finds that
analogous geographical distinctions characterize the material culture
of the past and reaches the conclusion that the relations of
environment to man and culture 1 The present paper is extracted
from a work now in course of preparation which is intended to bring
together in comprehensive form the antiquities of the continent; it is
thus not complete in itself. The several areas are tentatively outlined
to facilitate descriptive and comparative studies of the numerous
classes of artifacts; and the brief sketches here presented are
intended to familiarize the reader and student with the field as a
whole and with the relative culture status of its more important
subdivisions.
 _|.I4 AMERICAN ANTHROPOLOGIST [n. s., i6, 1914 must
play an important part in tiic prosecution of his researches and in the
analysis of aboriginal history. In the practical work of museum
classification and arrangement— a work which has served in part to
give form to this writing— archeological materials are necessarily
grouped primarily by continents and other natural divisions, and
secondarily by political divisions, such as states and territories.
Separation by the larger natural divisions is always necessary, but
separation by ethnic areas, or areas of culture characterization, as
they are sometimes called, is most advantageous. These areas may
be large or small according to the understanding or the needs of the
student. By their means he approximates the real or natural
grouping of the material traces of human achievement and studies
to advantage culture and culture relationships and the causes of the
resemblances and differences everywhere met with. The
geographical limitations of culture units are, as a matter of course,
not usually well defined. Cultures are bound to overlap and blend
along the borders and more especially along lines of ready
communication. But notwithstanding this, certain characteristics of
achievement or groups of culture traits within each area will be
found to separate it from its neighbors and afford effective means of
comparison with other culture groups. In the present work, keeping
in view the archeological rather than the ethnological evidence, it is
convenient to recognize eleven areas north of Mexico (pi. xxxii),
namely: (i) The North Atlantic area; (2) The Georgia-Florida area;
(3) The Middle and Lower Mississippi Valley Region; (4) The Upper
Mississippi and Lakes Region; (5) The Plains and Rocky Mountains;
(6) The Arid Region; (7) The California Area; (8) The Columbia-
Fraser Area; (9) The Northwest Coast Area; (10) The Arctic Coastal
Area; (11) The Great Northern-Central Area. To these may be added
(12) The Hawaiian Islands; and (13) The West Indies. These areas
are here made as few and simple as possible to avoid too great
complexity in conducting comparative studies of the several classes
of antiquities. The Middle and South American areas, also outlined
on the broadest possible plan, are as follows: (i) Northern Mexico;
(2)
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!

textbookfull.com