Pre-publication Copy

Table of Contents, Preface, and Chapters 1 through 3

Please realize that these pages have not been through the final editing process and are
still undergoing final revisions and editing and may contain typographical errors. Any
errors will be corrected before the book goes to press.

No part of this publication can be reproduced, stored in a retrieval system, or

transmitted, in any form or by any means, electronic, mechanical, photocopying or
otherwise, without the prior written permission of the publisher.
 Rev.Confirming Pages

Communication Research
ASKING QUESTIONS, FINDING ANSWERS

THIRD EDITION

Joann Keyton
North Carolina State University

key06767_fm_i-xxiv.indd i 1/29/10 3:31:19 PM

 Rev.Confirming Pages

COMMUNICATION RESEARCH, THIRD EDITION

Published by McGraw-Hill, a business unit of The McGraw-Hill Companies, Inc., 1221 Avenue of the
Americas, New York, NY 10020. Copyright © 2011 by The McGraw-Hill Companies, Inc. All rights reserved.
Previous editions 2006 and 2001. No part of this publication may be reproduced or distributed in any form or
by any means, or stored in a database or retrieval system, without the prior written consent of The McGraw-
Hill Companies, Inc., including, but not limited to, in any network or other electronic storage or transmission,
or broadcast for distance learning.
Some ancillaries, including electronic and print components, may not be available to customers outside the
United States.
This book is printed on acid-free paper.
1 2 3 4 5 6 7 8 9 0 DOC/DOC 1 0 9 8 7 6 5 4 3 2 1 0
ISBN 978-0-07-340676-3
MHID 0-07-340676-7
Vice President & Editor-in-Chief: Mike Ryan
Vice President SEM, EDP / Central Publishing Services: Kimberly Meriwether David
Executive Editor: Katie Stevens
Marketing Manager: Pam S. Cooper
Managing Editor: Meghan Campbell
Project Manager: Erin Melloy
Design Coordinator: Brenda A. Rolwes
Cover Designer: Studio Montage, St. Louis, MO
Photo Research Coordinator: Brian Pecko
Cover Image Credit: Getty Images
Production Supervisor: Nicole Baumgartner
Compositor: Laserwords Private Limited
Typeface: 9.5/12 Palatino
Printer: R. R. Donnelley

All credits appearing on page or at the end of the book are considered to be an extension of the copyright
page.
Library of Congress Cataloging-in-Publication Data
Keyton, Joann.
Communication research asking questions, finding answers / Joann Keyton.
—3rd ed.
p. cm.
ISBN 978-0-07-340676-3 (alk. paper)
1. Communication—Research. I. Title.
P91.3.K49 2010
302.207’2—dc22 2010001532

www.mhhe.com

key06767_fm_i-xxiv.indd ii 1/29/10 3:31:19 PM

 Rev.Confirming Pages

CONTENTS

Preface xviii

SECTION 1: RESEARCH BASICS

CHAPTER 1 Introduction to Communication Research 1

CHAPTER CHECKLIST 1
WHAT IS RESEARCH? 2
Your Relationship with Research 3

SCHOLARLY RESEARCH 4
Goals of Research 5
Research and Theory 7

COMMUNICATION AS A SOCIAL SCIENCE 7

The Scientific Approach 9
Characteristics of Science 11
Methodological Extremes 13

WHAT KINDS OF QUESTIONS DO COMMUNICATION SCHOLARS ASK? 13

The Nature of the Questions 14

SUMMARY 16
KEY TERMS 17

BOXED FEATURES
AN ETHICAL ISSUE : Is Communication Public or Private? 4
TRY THIS ! Evaluating Communication Questions 15

CHAPTER 2 The Research Process: Getting Started 18

CHAPTER CHECKLIST 18
THE RESEARCH PROCESS MODEL 19
IDENTIFYING THE RESEARCH PROBLEM 22
Formulating Research Ideas 22
Turning Topics into Preliminary Questions 22
Evaluating Your Questions 24 iii

key06767_fm_i-xxiv.indd iii 1/29/10 3:31:19 PM

 Rev.Confirming Pages

iv CONTENTS

USING LIBRARY AND DATABASE RESOURCES 24

Scholarly Journals 24
Books 25
Online Resources 26
Analyzing Your Resources 26
Adjusting the Question 28

USING AND ORGANIZING YOUR RESOURCES 29

Tracking Others’ References 29
Organizing What You Have Found 30
Summarizing What You Have Found 30

CONSIDERING THEORY IN RESEARCH 30

Developing Theory 31
Utility of Theory–Research Link 33

SUMMARY 34
KEY TERMS 34

BOXED FEATURES
TRY THIS ! Developing Initial Questions to Guide the Research Process 23
TRY THIS ! Searching for Source 27
AN ETHICAL ISSUE : Using the Ideas of Others 28
DESIGN CHECK : Evaluating the Literature You Found 29
TRY THIS ! Finding Theory in Journal Articles 33

CHAPTER 3 Introduction to Quantitative Research 35

CHAPTER CHECKLIST 35
WHAT IS QUANTITATIVE RESEARCH? 36
Examples of Quantitative Research 36
Deductive Reasoning 38
A Model for Conceptualizing Quantitative Research 38

CREATING THE FOUNDATION FOR QUANTITATIVE RESEARCH 41

RESEARCH HYPOTHESES FOR QUANTITATIVE RESEARCH 43
Directional Hypotheses 43
Nondirectional Hypotheses 43
Assessing Hypotheses 44
Null Hypotheses 44
Research Traditions in the Use of Hypotheses 44

key06767_fm_i-xxiv.indd iv 1/29/10 3:31:19 PM

 Rev.Confirming Pages

CONTENTS v

RESEARCH QUESTIONS IN QUANTITATIVE RESEARCH 45

TYPES OF VARIABLES 45
Independent Variables 46
Dependent Variables 48
The Relationship Between Independent and Dependent Variables 48

OPERATIONALIZING VARIABLES 50
MAKING THE CASE FOR QUANTITATIVE RESEARCH 51
Advantages of Quantitative Research 51
Limitations of Quantitative Research 52

ISSUES OF RELIABILITY AND VALIDITY 53

Threats to Reliability and Validity 53

SUMMARY 55
KEY TERMS 56

BOXED FEATURES
AN ETHICAL ISSUE : Giving Your Children Permission to Participate in a Research Study 37
DESIGN CHECK : What Is Sex? What Is Gender? 47
TRY THIS ! Identifying Independent and Dependent Variables 50
TRY THIS ! Operationalizing Variables 52

CHAPTER 4 Introduction to Qualitative Research 57

CHAPTER CHECKLIST 57
WHAT IS QUALITATIVE RESEARCH? 58
Examples of Qualitative Research 59
Inductive Analysis 62
A Model for Qualitative Research 62
Issues of Credibility in Qualitative Research 65

CONCEPTUALIZING RESEARCH QUESTIONS FOR QUALITATIVE RESEARCH 66

Assessing Research Questions 67

WHAT COUNTS AS DATA IN QUALITATIVE RESEARCH? 68

Interpreting Meaning 68
Level of Evidence 69

MAKING THE CASE FOR QUALITATIVE RESEARCH 69

Advantages of Qualitative Research 69
Limitations of Qualitative Research 69
Threats to Credibility in Qualitative Research 70

key06767_fm_i-xxiv.indd v 1/29/10 3:31:19 PM

 Rev.Confirming Pages

vi CONTENTS

COMPARING QUANTITATIVE AND QUALITATIVE RESEARCH 70

Key Differences in Quantitative and Qualitative Research 71
Key Similarities in Quantitative and Qualitative Research 71
Which Do I Choose? 72

SUMMARY 73
KEY TERMS 74

BOXED FEATURES
TRY THIS ! Identifying Contexts for Qualitative Research 61
DESIGN CHECK : What If Research Questions Are Not Presented? 67
AN ETHICAL ISSUE : Is Anything Off-Limits? 73

CHAPTER 5 Research Ethics 75

CHAPTER CHECKLIST 75
ETHICAL QUESTIONS IN PLANNING RESEARCH 76
Institutional Review Board 81
Informed Consent 82
Informed Consent and Quantitative Research 86
Informed Consent and Qualitative Research 86

ETHICAL ISSUES IN CONDUCTING RESEARCH 87

Intentional Deception 87
Using Confederates 88
Physical and Psychological Harm 89
Upholding Anonymity and Confidentiality 90
Videotaping and Audiotaping Participants 92
Debriefing Participants 93

ETHICAL ISSUES IN REPORTING RESEARCH 93

Ensuring Accuracy 93
Avoiding Plagiarism 94
Protecting the Identities of Participants 94

SUMMARY 95
KEY TERMS 96

BOXED FEATURES
AN ETHICAL ISSUE : Professional Association Guidelines for Conducting Research 78
AN ETHICAL ISSUE : Do Research Participants Have Any Ethical Responsibilities? 82
DESIGN CHECK : Do You Need Informed Consent for Your Research Project? 83
AN ETHICAL ISSUE : Would You Participate? 85
DESIGN CHECK : Private or Public? 87
TRY THIS ! What Would You Include? 92
AN ETHICAL ISSUE : Ethics in Proprietary Research 95

key06767_fm_i-xxiv.indd vi 1/29/10 3:31:19 PM

 Rev.Confirming Pages

CONTENTS vii

SECTION 2: QUANTITATIVE COMMUNICATION RESEARCH

CHAPTER 6 Measurement 97
CHAPTER CHECKLIST 97
MEASUREMENT PRINCIPLES 98
LEVELS OF MEASUREMENT IN COMMUNICATION RESEARCH 100
Discrete Data 100
Continuous Level Data 102
Ordinal Data 103
Interval Data 103
Ratio Data 106

ISSUES OF VALIDITY AND RELIABILITY 107

Validity 107
Face Validity 108
Content Validity 108
Criterion-Related Validity 108
Construct Validity 109
Reliability 110
Internal Reliability 110
Reliability Between Coders 112
Test–Retest Reliability 113
Split-Half Reliability 114
Improving Reliability 114
The Relationship Between Validity and Reliability 114
Threats to Validity and Reliability 115
Issues of Data Collection 115
Issues of Sample Representativeness 116
Alternative Explanations 117

ISSUES OF DATA INTERPRETATION 117

SUMMARY 118
KEY TERMS 119

BOXED FEATURES
AN ETHICAL ISSUE : What Am I Willing to Reveal About Myself ? 102
TRY THIS ! Changing Levels of Measurement 105
DESIGN CHECK : Questioning Measurement Techniques 106
DESIGN CHECK : Will These Results Apply to Me? 116

CHAPTER 7 Populations, Samples, and Sample Size 120

CHAPTER CHECKLIST 120
POPULATION AND SAMPLE 121
Defining the Population 122

key06767_fm_i-xxiv.indd vii 1/29/10 3:31:19 PM

 Rev.Confirming Pages

viii CONTENTS

ADDRESSING GENERALIZABILITY 123

Probability Sampling 124
Simple Random Sampling 126
Systematic Sampling 126
Stratified Random Sampling 127
Cluster Sampling 127
Nonprobability Sampling 128
Convenience Sampling 129
Volunteer Sampling 129
Inclusion and Exclusion Criteria 130
Snowball Sampling 131
Purposive Sampling 131
Quota Sampling 132
Sample Size 132
A Final Word on Sampling 134

SUMMARY 135
KEY TERMS 135

BOXED FEATURES
AN ETHICAL ISSUE : Participant Selection Can Make a Difference 128
DESIGN CHECK : Are Students Too Convenient? 130
DESIGN CHECK : Justifying Nonprobability Sampling 131
TRY THIS ! Identifying Populations, Generating Samples 133
AN ETHICAL ISSUE : Selected, But Not Required to Participate 134

CHAPTER 8 Quantitative Research Designs 136

CHAPTER CHECKLIST 136
THE EXPERIMENTAL FRAMEWORK 137
The Classical Experiment 139
Random Assignment of Participants 140
Creating Treatment and Control Groups 140
Manipulation Checks 142
Types of Experimental Design 143
Posttest Only 143
Pretest–Posttest 145
Factorial Design 145
Longitudinal Designs 147
Strengths of Experimentation 148
Limitations of Experimentation 149

QUASI-EXPERIMENTS 150
Field Experiments 152
Strengths and Limitations of Quasi-Experimental Research Designs 154

DESCRIPTIVE DESIGNS 154

Strengths and Limitations of Descriptive Research Designs 155

key06767_fm_i-xxiv.indd viii 1/29/10 3:31:19 PM

 Rev.Confirming Pages

CONTENTS ix

RESEARCHER EFFECTS AND PROCEDURAL BIAS 156

COMPARING RESEARCH DESIGNS 157
SUMMARY 158
KEY TERMS 158

BOXED FEATURES
AN ETHICAL ISSUE : Would You Ask Participants To . . . ? 142
TRY THIS ! Designing an Experiment 149
DESIGN CHECK : Is One Study Ever Enough? 153

CHAPTER 9 Surveys and Questionnaires 160

CHAPTER CHECKLIST 160
WHAT IS A SURVEY? 161
Types of Surveys 161
Online Surveys 162
Comparing Survey Types 164

DESIGNING A SURVEY 164

Evaluating Existing Questionnaires or Surveys 166
Writing Your Own Questionnaire 166
Designing Survey Items 167
Closed Questions 168
Open Questions 172
Choosing Between Open and Closed Questions 174
Designing the Flow of the Survey 175
How the Survey Looks 175

PRETESTING THE SURVEY OR QUESTIONNAIRE 175

SAMPLING ISSUES FOR SURVEYS 180
Response Rate 180

SURVEY RELIABILITY AND VALIDITY 181

ANALYZING AND REPORTING SURVEY DATA 182
SUMMARY 183
KEY TERMS 183

BOXED FEATURES
DESIGN CHECK : Are All Polls the Same? 164
TRY THIS ! Survey Design 166
DESIGN CHECK : Racial and Ethnic Group Identification 170
TRY THIS ! How Different Are Usually, Sometimes, and Seldom? 174
TRY THIS ! Does This Questionnaire Need Modification? 178
AN ETHICAL ISSUE : Would You Participate? 180

key06767_fm_i-xxiv.indd ix 1/29/10 3:31:19 PM

 Rev.Confirming Pages

x CONTENTS

CHAPTER 10 Descriptive Statistics, Significance Levels, and Hypothesis Testing 184

CHAPTER CHECKLIST 184
NUMBERS INTO STATISTICS 185
NORMAL CURVE 186
Skewed Distributions 186
Distributions of Data 187

DESCRIPTIVE STATISTICS 189

Number of Cases 189
Measures of Central Tendency 189
Mean 190
Median 190
Mode 190
Measures of Dispersion 191
Range 192
Standard Deviation 192

APPLICATION OF DESCRIPTIVE STATISTICS 193

Frequencies 193
Percentages 194

CRUNCHING THE NUMBERS: CALCULATOR OR COMPUTER? 195

SIGNIFICANCE LEVELS 196
HYPOTHESIS TESTING 198
A FINAL WORD 200
SUMMARY 200
KEY TERMS 201

BOXED FEATURES
TRY THIS ! Are the Data Normal or Skewed? 190
AN ETHICAL ISSUE : Mistakes in Calculations 194
DESIGN CHECK : Describing Variables 195

CHAPTER 11 Testing for Differences 202

CHAPTER CHECKLIST 202
INFERENTIAL STATISTICS 203
Alternative and Null Hypotheses 204
Degrees of Freedom 204
Four Analytical Steps 205

CHI-SQUARE 206
One-Dimensional Chi-Square 206
Contingency Analysis 207
Limitations of Chi-Square 208

key06767_fm_i-xxiv.indd x 1/29/10 3:31:19 PM

 Rev.Confirming Pages

CONTENTS xi

THE t-TEST 209

Common Forms of t-Test 210
Two-Tailed t-Test 211
One-Tailed t-Test 212
Limitations of the t-Test 212

ANALYSIS OF VARIANCE 212

ANOVA Basics 213
One-Way ANOVA 215
Two-Way ANOVA 216
Factorial ANOVA 218
Limitations of ANOVA 219

ASKING FOR STATISTICAL HELP 219

SUMMARY 220
KEY TERMS 220

BOXED FEATURES
DESIGN CHECK : Interpreting Chi-Squares 209
DESIGN CHECK : Interpreting t-Tests 211
AN ETHICAL ISSUE : Are Two Categories Fully Representative? 213
TRY THIS ! Should Differences Be Significant or Practical? 217
DESIGN CHECK : Interpreting F 218

CHAPTER 12 Testing for Relationships 222

CHAPTER CHECKLIST 222
BASIC ASSUMPTIONS 224
Alternative and Null Hypotheses 224
Degrees of Freedom 225
Four Analytical Steps 225

CORRELATION 225
Interpreting the Coefficient 226
Amount of Shared Variance 227
Plotting the Data 228
Examples of Correlation 228
Other Forms of Correlation 230
Limitations of Correlation 230

REGRESSION 231
Linear Regression 232
Multiple Regression 233
Interpreting Multiple Regression 233
Regression Examples 235
Other Forms of Regression 238
Limitations of Regression 238

key06767_fm_i-xxiv.indd xi 1/29/10 3:31:19 PM

 Rev.Confirming Pages

xii CONTENTS

STRUCTURAL EQUATION MODELING 238

SEM Example 239
Other Forms of SEM 240

CAUTIONS IN USING STATISTICS 240

SUMMARY 241
KEY TERMS 242

BOXED FEATURES
DESIGN CHECK : Paying Attention to Details 224
DESIGN CHECK : Interpreting Correlations 231
DESIGN CHECK : Interpreting Multiple Regression 236
TRY THIS ! Identifying Independent Variables for Multiple Regression 237

CHAPTER 13 Quantitative Analysis of Text 243

CHAPTER CHECKLIST 243
CONTENT ANALYSIS 244
What Content Can Be Analyzed? 246
The Content Analysis Process 246
Selecting What to Code 246
Developing Content Categories 247
Units of Analysis 248
Training Coders 249
Coding Reliability 249
Validity 251
Interpreting the Coding Results 251
Computers and Content Analysis 252
Strengths of Content Analysis Research 253
Limitations of Content Analysis Research 253

INTERACTION ANALYSIS 254

Gathering and Preparing the Interaction for Coding 254
Coding the Interaction 256
Analyzing and Interpreting the Coding 256
Strengths of Interaction Analysis 256
Limitations of Interaction Analysis 256

DESIGN CONSIDERATIONS 257

SUMMARY 257
KEY TERMS 258

BOXED FEATURES
DESIGN CHECK : Content Analyzing Media Messages 246
TRY THIS ! Identifying Themes of Freshman Socialization 250
DESIGN CHECK : How Did the Researcher Perform the Content Analysis? 251
AN ETHICAL ISSUE : Taking Content Out of Context 253

key06767_fm_i-xxiv.indd xii 1/29/10 3:31:20 PM

 Rev.Confirming Pages

CONTENTS xiii

SECTION 3: QUALITATIVE COMMUNICATION RESEARCH

CHAPTER 14 Designing Qualitative Research 259

CHAPTER CHECKLIST 259
RESEARCHER SKILLS FOR QUALITATIVE METHODOLOGIES 260
THE RESEARCHER’S ROLE IN QUALITATIVE METHODOLOGIES 261
Forms of Participant Observation 261
Complete Participant 261
Participant as Observer 262
Observer as Participant 262
Complete Observer 263
Moving Between Researcher Role 264

IDENTIFYING THE RESEARCH FOCUS 264

Consulting the Literature 266
Concept Map 266
Research Questions and Objectives 266

SAMPLING IN QUALITATIVE DESIGNS 268

Special Considerations for Sampling in Qualitative Research 269

GAINING ACCESS 270

Becoming Familiar with People and Places 272
Developing Trust 272
Developing Rapport 272
Locating Key Informants 273
Stumbling Onto Something Interesting 273

DESIGNING A QUALITATIVE RESEARCH PROJECT 274

What Constitutes Data in Qualitative Research? 275
Your Impact as a Researcher 275

FINALIZING THE RESEARCH DESIGN 277

SUMMARY 277
KEY TERMS 278

BOXED FEATURES
AN ETHICAL ISSUE : One Role? Or Two? 263
TRY THIS ! Assuming Researcher Roles 264
DESIGN CHECK : Why Was the Researcher There? 270
DESIGN CHECK : Are the Data Credible? 276

CHAPTER 15 Qualitative Methods of Data Collection 279

CHAPTER CHECKLIST 279
COLLECTING QUALITATIVE DATA 280

key06767_fm_i-xxiv.indd xiii 1/29/10 3:31:20 PM

 Rev.Confirming Pages

xiv CONTENTS

Observation Strategies 280

The Seamless Container 280
Ask Questions to Aid Observation 281
Taking Notes 281
Transcribing Notes 283
What if You Cannot Take Notes 283
Describing and Analyzing 283

FIELD INTERVIEWING 284

Electronic Interviewing 286
The Interview Process 286
Conceptualizing the Interview Study 286
Designing the Interview 287
Conducting the Interview 288
Asking Questions 289
Listening 291
Concluding the Interview 291
Transcribing the Interview 291
Strengths and Limitations of Interview Research 291

FOCUS GROUPS 292

Planning Focus Group Research 293
The Focus Group Outline 293
Selecting Paticipants 294
Conducting Focus Group Research 295
Guiding and Maintaining the Conversation 295
Data from Focus Group Discussions 295
Focus Group Strengths and Limitations 295

COLLECTING NARRATIVES 297

Sources for Stories 298
Strengths and Limitations of Narrative Research 299

ETHNOGRAPHY 300
Entering the Scene 301
Strengths and Limitations of Ethnographic Research 302

SUMMARY 302
KEY TERMS 303

BOXED FEATURES
TRY THIS ! Your Classroom as a Seamless Container 281
DESIGN CHECK : Where Did the Interviews Occur? 287
TRY THIS ! Motivating Attendance at Focus Groups 296
AN ETHICAL ISSUE : Research or Selling? 298
DESIGN CHECK The Difference Between the Research Question and Asking Questions 299
DESIGN CHECK : Reading Ethnography 302

key06767_fm_i-xxiv.indd xiv 1/29/10 3:31:20 PM

 Rev.Confirming Pages

CONTENTS xv

CHAPTER 16 Analyzing Qualitative Data 304

CHAPTER CHECKLIST 304
AN OVERVIEW 305
CHOOSING AN ANALYTIC METHOD 305
THE PROCESS OF ANALYSIS 306
Analytical Memos 306
Diagramming the Data 307
Coding and Analyzing with Software 308
Coding and Categorizing Data 309

GROUNDED THEORY 311

THEMATIC ANALYSIS 313
THE PROCESS OF INTERPRETATION 314
Evaluating Interpretation 315
Participant Quotes 315
Credibility 315
Triangulation 316

A FINAL WORD 317

SUMMARY 317
KEY TERMS 318

BOXED FEATURES
DESIGN CHECK : How Are Participant Quotes Used? 315

SECTION 4: READING AND WRITING RESEARCH REPORTS

CHAPTER 17 Reading and Writing the Quantitative Research Report 319

CHAPTER CHECKLIST 319
THE LITERATURE REVIEW 320
The Problem Statement 321
Types of Material to Include in the Literature Review 321
Organizing the Literature Review 321
Presenting the Research Questions and Hypotheses 322

THE METHODS SECTION 323

Describing Participants 323
Describing the Research Procedure 323
Describing the Variables 324

THE RESULTS SECTION 325

Using Tables and Graphs 325

key06767_fm_i-xxiv.indd xv 1/29/10 3:31:20 PM

 Rev.Confirming Pages

xvi CONTENTS

THE DISCUSSION SECTION 325

Developing Interpretations from Results 326
Manner of Presentation 326
Presenting the Limitations 328
Recommending Future Research 328

FINISHING THE QUANTITATIVE RESEARCH REPORT 328

Title 329
Title Page 329
Abstract 329
References 329

USING APA STYLE 329

Citing Others’ Work 329
Creating the Reference List 330

ISSUES IN WRITING 331

The Revision Process 332

SUMMARY 332
KEY TERMS 333

BOXED FEATURES
DESIGN CHECK : Do You Have the Basic Information? 324
AN ETHICAL ISSUE : Dealing with Unexpected Results 327
TRY THIS ! How to Read a Research Report 328
TRY THIS ! Submit Your Research Paper to a Communication Convention 330

CHAPTER 18 Reading and Writing the Qualitative Research Report 334

CHAPTER CHECKLIST 334
REPRESENTING QUALITATIVE RESEARCH 335
Beginning the Research Report 336
The Literature Review 336

WRITING ABOUT THE METHOD 337

Revealing the Identity of Participants 337

PRESENTING THE DATA 338

Authorial Voice 338
Adopting a Writing Style 339
The Writing Style of the Realist 339
The Writing Style of the Confessional 340
The Writing Style of the Impressionist 341
The Writing Style of the Critical Scholar 342

key06767_fm_i-xxiv.indd xvi 1/29/10 3:31:20 PM

 Rev.Confirming Pages

CONTENTS xvii

FINDINGS AND INTERPRETATION 344

Balancing Description and Analysis 344
Identifying and Presenting Organizing Patterns 344
Using Participants’ Quoted Material 346

DRAWING AND SUPPORTING CONCLUSIONS 346

Revisiting Your Analysis 347

FINISHING THE QUALITATIVE RESEARCH REPORT 348

Title 348
Abstract 348
Using the APA Style Manual 349

SUMMARY 349
KEY TERMS 350

BOXED FEATURES
AN ETHICAL ISSUE : Creating Something from Nothing 342
DESIGN CHECK : Who Did What? How? When? 343
TRY THIS ! Describe, Then Analyze 345
DESIGN CHECK : Letting Participants Speak 346

APPENDIXES

A Personal Report of Communication Apprehension 351

B Formulas and Steps for Statistical Calculations 353
C Random Numbers Table 370
D Sample Protocol for Experimental Study 373
Glossary G–1
References R–1
Author Index I–1
Subject Index I–5

key06767_fm_i-xxiv.indd xvii 1/29/10 3:31:20 PM

 Rev.Confirming Pages

PREFACE

This book began many years ago when I was how the organization defined and formalized the
forced to think about research methods in a relationship between managers and employees.
new way. Approached by an organization to Again, I could have stopped there, presented
help them assess their human resources policy the information to the organization, and con-
and training, I was overwhelmed by the task of sidered myself done with the task. But a new
identifying the problems and discovering and research opportunity presented itself—to test
developing practical, yet theoretically based, employee and management acceptance of the
solutions. I eagerly agreed to help the organi- policy and training I proposed. After designing
zation because their problem—how to create a and conducting survey and experimental studies,
workplace in which respect and dignity were I was able to tailor the messages the organization
core values—was an opportunity to test my wanted to send in a way that was acceptable to
skills and abilities as a communication research- and understandable by employees.
er. I knew that I first needed to develop content The point to this story is that research is a pro-
expertise. After conducting library and data- cess. In some cases, library research from the vast
base searches, I realized that the information I store of communication literature may answer
obtained was incomplete and at times conflict- our research questions. In other cases, research-
ing. Yet, this information allowed me to develop ers must design studies to collect quantitative or
an initial response that included policy, training, qualitative data that, when analyzed, will answer
and evaluation procedures. I could have stopped those questions. Too many times, I hear students
there, presented the information to the organiza- describe research as having little relevance to their
tion, and considered myself done with the task. lives. But, when the research process is designed
Yet, conducting research within the organization to answer questions about communication issues
was a unique research opportunity, allowing me that are important to students, it provides them
to both contribute to the organizational com- with a new perspective. Rather than considering
munication literature base and create a more research a laborious chore, they recognize that
tailored response for the organization. research is the most effective means for answering
Moving further into this process, I conducted meaningful questions in their lives. As important-
focus groups with employees at all levels to ly, many employers seek research skills in hiring
gain insight into their personal experiences in and promoting employees
the organization. Policy and training would not The other point to my story is that far too often,
be as effective if they did not address the needs I hear students and professors make claims such
of employees and management. Data from the as, “I’m a quantitative researcher” or “I’m an
focus groups revealed that employees identi- ethnographer.” In reality, researchers must have
fied a few managers whom they believed to a broad understanding and appreciation of all
be effective in dealing with employee issues. methodologies—quantitative and qualitative—
I conducted field interviews with these indi- to conduct their research effectively. The third
viduals. Data from those interviews revealed that edition of this book continues to emphasize three
inconsistencies among the organization’s current important points:
policies, training, and evaluation procedures were
a significant problem. Next, I conducted a content 1. All research starts with an initial research
analysis of the organization’s policies to examine question or problem.
xviii

key06767_fm_i-xxiv.indd xviii 1/29/10 3:31:20 PM

 Rev.Confirming Pages

PREFACE xix

2. Research is a process in which the researcher literature. Regardless of the role in which
makes important decisions at crucial points students use their research knowledge, they
about what to do and how to do it. This is must be able to read and understand the com-
in contrast to viewing research simply as a munication research literature.
series of steps to be completed. This book provides several features to help
3. To answer the varied nature of questions students succeed in both roles. First, hundreds
about communication, one must be familiar of examples are drawn from published com-
with both quantitative and qualitative meth- munication research to provide clear direction
odologies. on what this process or step looks like. These exam-
ples, 197 of which are new to the third edition,
Communication Research: Asking Questions, are drawn from recent journal articles, which
Finding Answers covers basic research issues are available in the Communication and Mass
and processes for both quantitative and quali- Media Complete database, or available online
tative approaches appropriate for communi- through university and college library subscrip-
cation students with no previous research tions. Using examples from the breadth of the
methods experience. The text’s guiding prin- discipline (for example, persuasion, interper-
ciple is that methodological choices are made sonal, group, organizational, mass communica-
from one’s research questions or hypotheses. tion, and public relations) lessens the ambiguity
This avoids the pitfall in which students learn between information presented in the book and
one methodology or one methodological skill students’ understanding and potential applica-
and then force that method to answer all types tion of the information.
of questions. Second, the book incorporates two kinds
of boxes, placed throughout the chapters that
alert students to the nuances of the research
FEATURES process. The first, Design Check, alerts stu-
dents to the practical and logistical issues that
The primary purpose of this textbook is to intro- student researchers should consider when
duce students to communication research meth- designing a study. These are the same issues
ods by meeting two objectives. The first objective that students should ask of the research studies
is to help students become better consumers of they read, as how these issues are addressed
the communication research literature by empha- by researchers influences study outcomes and
sizing effective methods for finding, consuming, data interpretations. The second type of box,
and analyzing communication research. This An Ethical Issue, alerts students to issues of
objective is important because students are con- research ethics and integrity. Not only must
sumers of the communication literature through researchers balance practical and logistical
their participation in communication courses. issues, they must do so while addressing
The second objective is to provide a path for stu- ethical issues that occur when people and their
dents who wish to develop and conduct research communication artifacts are used as the basis
projects. To those ends, this book provides cov- of research.
erage of the entire research process: how one Third, the book is based on active pedago-
conceptualizes a research idea, turns it into an gy and the philosophy that students learn best
interesting and researchable question, selects a by doing. Chapter Checklists begin each chapter
methodology, conducts the study, and writes up to highlight for students the essential learning
the study’s findings. I believe that students who objectives for each chapter. The objectives help
can effectively navigate, select, and use the com- students make discrete distinctions about the
munication research literature can become effec- research process and give students a standard
tive researchers, and, reciprocally, that students for what they should be able to demonstrate
engaged in communication research will be able after reading and studying chapter material.
to more effectively use the existing research Try this! boxes are placed throughout the chapters

key06767_fm_i-xxiv.indd xix 1/29/10 3:31:20 PM

 Rev.Confirming Pages

xx PREFACE

to engage students in short research activities that depth and more examples; and, this information
can be used in the classroom with individuals was reorganized for improved flow that better
or groups, or as short homework assignments. mirrors the qualitative research process.
End-of-chapter summaries have been developed The book remains divided into four sections.
as point-by-point summaries of information pre- In the first section, Research Basics, students are
sented in the chapter. Stated simply, these factual introduced to the research process, its basic prin-
statements can help direct students’ study of the ciples, both quantitative and qualitative research,
material and be used as a stimulus to extract and research ethics before specific methodologi-
students’ understanding and application of the cal techniques are addressed. This organization
material. Key terms are boldfaced within the text emphasizes that research is a process, not just
and listed at the end of chapter. Key term defini- one type of method or research skill. Chapters 1
tions can be found in the glossary at the end of and 2 are introductory to research in general and
the book. are neutral with respect to methodology. The
Finally, the book focuses on students. It is issues raised in these initial chapters are issues
written for them—to their level of knowledge that both quantitative and qualitative research-
and understanding about human communica- ers must address. Specifically, in Chapter 1, the
tion, the communication research literature, and discussion of communication research has been
the relative research processes. My goal in writ- enhanced to make clearer distinctions among
ing the chapters was to explain the research steps the differences and similarities among social
and identify the steps researchers take in devel- science, rhetorical, and critical research tradi-
oping and conducting communication research. tions. Additional information has been includ-
With study and instruction, students should be ed to show that communication research con-
able to use this material and integrate it with ducted under different perspectives influences
what they know and are familiar with from their what counts as data and how communication
other communication courses to accomplish two is explained. Chapter 2 was updated to reflect
objectives: 1) to be more analytical and make recent changes in web and online resources with
more sophisticated interpretations of the com- which students should be familiar.
munication research they read, and 2) to design Then Chapters 3 and 4 provide introductions to
and conduct basic quantitative and qualitative both quantitative and qualitative methodologies.
research studies. This arrangement encourages students to consid-
er both methodologies and to gain a foundation
in each before proceeding to detailed information
in subsequent chapters on how specific methods
WHAT’S NEW TO THE THIRD within each work. Chapter 3 was revised in two
EDITION AND ORGANIZATION ways. First, a clearer explanation of the rela-
OF THE TEXT tionship between independent and dependent
variables was added. Second, the information
The book presents a balance of quantitative and on intervening and confounding variables was
qualitative research because the communication removed. Both of these changes are intended to
research literature embraces both approaches. In make the central ideas of quantitative research
addition to the updating of published research easier for students. Chapter 4 was revised to
examples, the biggest change to the third edition is emphasis the centrality and importance of the
the increased focus on qualitative research. More iterative process to qualitative research.
examples are given, more direction is provided Chapter 5 is devoted to issues of research eth-
in how to design a qualitative study (Chapter 14), ics—issues students must consider regardless of
and collect (Chapter 15) and analyze qualitative which methodology they choose. Information in
data (Chapter 16), as well as writing a qualitative this chapter was checked with, and if necessary
research report (Chapter 18). Information in each updated to, current professional standards and
of these chapters was strengthened with greater federal government regulations and standards.

key06767_fm_i-xxiv.indd xx 1/29/10 3:31:20 PM

 Rev.Confirming Pages

PREFACE xxi

A Design Check box on research on the Internet The fourth section of the book, Reading
was added to reflect researchers’ increasing use and Writing Research Reports, provides sepa-
of this data collection technique and the issues rate chapters for concluding the research pro-
it raises. With the increasing use of technology, cess. These chapters demystify this stage of the
the issue of what’s private and what’s public has research process for students, whether they are
changed. Thus, the distinction between private reading the research literature or ready to write a
and public is now conceptualized as a continu- research report. Researchers relying on tradition
um rather than a dichotomy. and customary practices are able and adept con-
The second section of the book, Quantitative sumers of the research literature. Alternatively,
Communication Research, provides detailed students confront the research literature with
coverage on how research is conducted with little understanding of how and why finished
quantitative methodologies. Chapters 6 through research reports look the way they do. By having
9 explain measurement concepts, sampling pro- access to this insider information, students will be
cedures, and experimental, quasi-experimen- able to prepare their final projects in the tradi-
tal, descriptive, and survey research designs. tions of the discipline as well as be able to better
Chapters 8 and 9 have been changed the most. decode the research literature.
Chapter 8 includes additional coverage of the Across all of the chapters, I worked to be more
advantages and limitations of quasi-experi- inclusive of examples drawn from intercultural
ments, as this form of experimentation is preva- research, research by international scholars, and
lent in communication research. Still, the chapter research about technology or that used technol-
begins with description of the classic experiment, ogy in data collection. With the shift from APA
as that is the theoretical model for other experi- 5th to 6th, I chose to create all of the citations
mental designs. More attention to independent and references in APA 6th style, and have made
and dependent variables has been added earlier special note of this where appropriate. Finally, a
in the chapter. Chapter 9 on surveys and ques- note about online resources available to students
tionnaires has a new section that provides rec- was added to each chapter. Whether you assign
ommendation for online survey data collection. students to work with these resources inside or
Enhanced information on survey layout has also outside of class, or expect students to use these
been included. materials on their own, the online resources
On the advice of reviewers, all material on provide students with ways to enhance and test
significance levels and hypothesis testing has their knowledge of research methods.
been moved to Chapter 10 and its treatment of In talking with colleagues who also teach
descriptive statistics. A section on structural research methods, I have found that instructors
equation modeling was added to Chapter 12, differ greatly in their treatment of statistical
to mirror the increased use of this statistical test concepts. To accommodate these differences in
in communication research. A figure was also pedagogical style, this book presents the con-
added to help students distinguish between ceptual foundation of each test supported with
types of tests of relationships presented in this examples of the test from the research literature.
chapter. As in the second edition, each of the Each test is discussed from the point of view
statistical chapters emphasizes interpreting the of a student who finds this test in the litera-
results of these tests. Chapter 13 explores two ture. Questions used to develop these chapters
quantitative methods for analyzing text and include: How should I read the results of any
message content. particular test? How do I connect these results
Changes to the chapters in the third section, to the research questions and hypotheses the
Qualitative Communication Research, are noted authors proposed? Alternatively, for those stu-
above: In essence, more information, more recom- dents who want to design and conduct quanti-
mendations, more research examples were added, tative research projects, the statistical formulas
especially to the chapters on data collection and (with worked examples as models for each
data analysis. test) appear in an appendix; a sample dataset

key06767_fm_i-xxiv.indd xxi 1/29/10 3:31:20 PM

 Rev.Confirming Pages

xxii PREFACE

and instructions for SAS and PASW (or SPSS) about research methods and statistics. When
software appear online. contacted by e-mail, these colleagues were both
prompt and generous.
I have also benefited from the many under-
TEACHING SUPPLEMENTS graduate students in my research methods course
who continued to say that they did not understand
Detailed instructor’s materials are available at after I had explained a concept or technique. Their
the Instructor Center website accompanies the questioning and my inability to always provide
book (www.mhhe.com/keyton3). The pass- them an appropriate and acceptable answer pro-
word-protected instructor’s center includes vided the motivation for this text.
sample syllabi, teaching tips, chapter and course This third edition has benefited from the many
assignments, exercises and PowerPoint outlines instructors and students who have e-mailed me
for each chapter, a test bank, and worksheets with questions or issues they would like me
for each chapter. Typically one or two pages in to address or explain further. I appreciate this
length, worksheets can be used as a homework or feedback-in-progress and much of it has been
in-class assignment for students to review their incorporated here.
knowledge and understanding about the mate- I also thank the scholars who reviewed this
rial presented. Question types include objective text during its development for the encourage-
(for example, fill in the blank), comprehension ment and wisdom they extended. Reviewers for
(for example, explain how academic research the third edition were: Tammy Swenson Lepper,
differs from proprietary research), and behav- Winona State University; Mark Comadena, Illinois
ioral (for example, given a set of variables the State University; Nancy Curtin-Alwardt, Millikin
student is asked to write research questions and University; Jong Kang, Illinois State University;
hypotheses). For those chapters that cover statis- Paaige Turner, Saint Louis University; Deden
tics or the analysis and interpretation of qualita- Rukmaan, Savannah State University; Darin
tive data, additional worksheets are available, Garard, Santa Barbara City College; Robert Brown,
which provide students with the opportunity to Salem State College; Steven Venette, University
work several examples from raw data through to of Southern Mississippi; Bo Feng, University of
interpretation. California, Davis; Rita Rahoi-Gilchrest, Winona
The Student Center can also be found at the State University; Lindsey Chamberlain, Ohio
same website (www.mhhe.com/keyton3). This State University; Suzanne Piotrowski, Rutgers
site is not password-protected and includes University, Newark; Xin Ren, California State
PowerPoint outlines for each chapter; URLs University, Sacramento.
for websites that support or provide addi- Thanks to the McGraw-Hill team, led by
tional material presented in the chapters; online executive editor Katie Stevens. Others on this
review tests; and a glossary. Sample datasets team included Craig Leonard, developmental
and instructions for using SAS and PASW (or editor; Meghan Campbell, managing editor; Erin
SPSS) for the descriptive and inferential statis- Melloy, project manager; and Pam Cooper, mar-
tics presented in Chapters 10, 11, and 12 are also keting manager. They helped me produce the
included here. finished product.
In the first edition, I thanked my colleagues—
Tommy Darwin, Steve Rhodes, and Pradeep
ACKNOWLEDGMENTS Sopory. In the second edition, I added Ron
Warren, Debbie Ford, and Tracy Russo. Each
In writing this book, I have benefited from the of these six people have enriched and chal-
generosity of researchers, scientists, and schol- lenged my role as researcher. For this third edi-
ars from many disciplines around the world. tion, I’d like to thank Paul Schrodt, Ryan Bisel,
Unlike many other bodies of knowledge, the Stephenson Beck, and Renee Meyers for always
Web has become a cornucopia of information returning the e-mails in which I proposed a

key06767_fm_i-xxiv.indd xxii 1/29/10 3:31:20 PM

 Rev.Confirming Pages

PREFACE xxiii

methods question or conundrum. The book has for your special friendship. The completion of this
benefited from those online discussions. revision is another reason to enjoy two of the finer
I would also like to thank the many under- things in life.
graduate and graduate students who have This revision came at a difficult time. Sally
worked with me at North Carolina State had passed during my beginning work on this
University, University of Kansas, and University edition. Sonny and Cher are making sure that
of Memphis on research projects and who have I live up to my promise that I will not forget
worked through research issues (and chal- what my real job is: to let the dogs in, let the dogs
lenged me) in methodology classes. For me, out, let the dogs in, let the dogs out. . . .
methodology is the best teaching assignment I Jeff—this book is for you. As a student, you
can have. would not allow me to let you down. As a friend,
Thank you to my friends—Bob, Randy, you have not let me down. Your invaluable les-
Christie, Sherry, Joanne and Tom, Pat and Jay, sons, both professional and personal, helped me
Linda, Liz and Kim, Allison, Amy, Stephanie, write this text in the beginning and through its
Kelby, Dave, Christi and Allen, and Bill. I appreci- two revisions. Thanks for your continual sup-
ate your friendship and support. Again, fs, thanks port and encouragement. My love always.

key06767_fm_i-xxiv.indd xxiii 1/29/10 3:31:20 PM

 Rev.Confirming Pages

xxiv PREFACE

Letter to Students
Welcome to communication research methods. The principles, traditions, and norms presented
in this text are simply that: principles, traditions, and norms. This means that these charac-
teristic ways of conducting and writing about research and evaluating research are not fixed.
Although enduring, these principles, traditions, and norms can and are being changed.
New statistical procedures are being developed (or older statistical procedures are being
reintroduced) by researchers to help deal with the complexity of and connections among com-
munication concepts. Researchers are exploring alternative ways for presenting qualitative
data. When a researcher uses a method or technique of collecting, analyzing, or presenting
data that is unique or different, the writer should provide arguments for its use and its valid-
ity or credibility. I think that’s fair.
As you explore research methods, I recommend that you first take on the role of being a
consumer of scholarly research. Undoubtedly, your instructors have asked (and will con-
tinue to ask) you to read articles and book chapters of scholarly research that present data
about communication phenomena. With an understanding of research methods, the data and
authors’ interpretations will be more meaningful. As importantly, you will begin to appreci-
ate how you can use research methods skills in your workplace and the community.
I also encourage you to resist the urge to label researchers or yourself as quantitative
researchers or qualitative researchers. Communication is such a diverse and complex
phenomenon that both quantitative and qualitative methods are needed for its exploration.
Without a doubt, many researchers favor quantitative over qualitative, or qualitative over
quantitative; it’s likely they have developed strengths in one area. And that’s okay! Even if
you choose to favor one type of research method over another, you’ll need to have enough
skills to read a variety of research reports.
Miller (2000) reminds us: Our goal as communication researchers should be to pursue
important questions about communication rather than categorizing ourselves as a certain
type of researcher. It is only possible to learn from the breadth and depth of communication
research if we can read and understand it.
Learning research methods is much like learning another language and culture. It’s not
likely that you will become proficient in Spanish, for example, at the end of one course.
Likewise, to be proficient you need to immerse yourself in reading about and the doing of
research methods. The authors of articles in scholarly books and chapters have had more
than one course in research methods. It’s a journey, not a location. Even if you do not take
another course in research methods, you are learning about research methods as you read
articles and book chapters and as you question “why did they do that?” and “how did they
get to that conclusion?”
I hope this text helps in your journey.

Best,

Joann

Miller, K. I. (2000). Common ground from the post-positivist perspective: From “straw person” argument to collabora-
tive coexistence. In S. R. Corman & M. S. Poole (Eds.), Perspectives on organizational communication: Finding common ground
(pp. 46–67). New York: Guilford Press.

key06767_fm_i-xxiv.indd xxiv 1/29/10 3:31:20 PM

 Rev.Confirming Pages

CHAPTER ONE

Introduction to Communication
Research

Chapter Checklist

After reading this chapter, you should be able to: 6. Differentiate among the characteristics of
1. Identify instances in which you could use or science.
conduct communication research as a stu- 7. Distinguish between research question and
dent, use or conduct communication research hypothesis.
as a professional, and use the results of com- 8. Describe the differences among questions of
munication research in your personal life. fact, variable relations, value, and policy.
2. Explain the goals of research. 9. Identify questions about communication that
3. Explain the relationship of research and you believe are worth pursuing.
theory.
4. Explain communication research as a social
science.
5. Describe how communication research from
a social science perspective is different from
other forms of communication research and
other forms of social science research.

key06767_ch01_001-017.indd 1 1/25/10 7:15:16 PM

 Rev.Confirming Pages

2 CHAPTER 1 / INTRODUCTION TO COMMUNICATION RESEARCH

As a student in a research methods course, you relying on the results of research performed by
have two roles. In one role, you are a consumer others, on the detective work of others. It would
of communication research. You read summaries be impossible for you to answer your set of ques-
of research in your textbooks. In some courses, tions without such input. For example, for the
your instructors may require you to read and question “What is the annual income of alumni
analyze research articles published in the disci- with my major?” it would not be realistic for
pline’s journals. you to survey graduates in your major field to
In the other role, you are a researcher collect- discover their annual income. More likely you
ing and interpreting data to answer research relied on a survey conducted by a professional
questions and hypotheses. These activities may association, an alumni association, or a fraternity
be part of the course for which you are read- or sorority. You used the reported results of their
ing this book, an independent study, an upper- work to answer your question. Although some-
division course, or your senior project. The one else was doing the research, you still needed
information in this book can help you succeed to evaluate the efficacy of their research to gauge
in both roles. But before you identify yourself its usefulness in answering the question.
with either or both roles, turn your attention to You are also familiar with other types of
answering the question “What is research?” research. News reports profile the results of
research each day. You have heard the results
of medical research reported in the news. Dur-
WHAT IS RESEARCH? ing political campaigns, the results of preference
polls are reported in the news and archived on
In its most basic form, research is the process news organization websites. And, no doubt,
of asking questions and finding answers. You you have heard the results of research on under-
have likely conducted research of your own, age drinking and drug use. If you work, your
even if it wasn’t in the formal sense. For exam- company may have conducted research on the
ple, as you chose which college or university preferences of its customers or the quality of its
to attend, you asked questions of students, fac- products.
ulty, and staff at the various institutions you The point here is that research is all around
were considering. You might also have looked us, often presented in ways that we would not
on web pages for the different colleges and recognize as research. Thus, research, as we will
universities for answers to your questions or study it, is the discovery of answers to questions
used the survey results from U.S. News & World through the application of scientific and system-
Report, which ranks America’s colleges and atic procedures. Given this basic definition of
universities. As you made choices about your research, you can see that you probably come
major, you read the college bulletin, talked to into contact with several forms of research on
students and an advisor, and perhaps even a daily basis. You probably also use the results
talked to professionals in the field you believed of research in making both personal and profes-
you wanted to pursue. In these activities, you sional decisions.
sought answers to your questions. Which The specific focus of this text is communica-
school is best for me? Which school has the type tion research—that is, quantitative or qualitative
of student experience I am looking for? Which research conducted by communication scholars
schools are affordable? What is the annual about communication phenomena. The focus
income of alumni with my major? What kinds is also on research conducted from a social sci-
of career opportunities can I expect? By asking ence perspective, which is distinct from rhe-
these questions, you were taking on the role of torical research and also distinct from critical
detective as you tracked down the information research. Yet, distinctions among these three
needed to make a decision. perspectives—social science, rhetorical, and
Not only were you asking questions and critical—are not always clear (Craig, 1993), and
seeking answers, but more than likely you were scholars working from the other perspectives

key06767_ch01_001-017.indd 2 1/25/10 7:15:16 PM

 Rev.Confirming Pages

WHAT IS RESEARCH? 3

do use some methods more commonly associ- as well as jobs in political, organizational, and
ated with social science research. Social science legal communication, are just a few in which
research is conducted through the use of sci- research plays a central role in decision making.
entific and systematic methods, and it is based Even though their organizational title may not
on the assumption that research can uncover be “researcher,” many employees at the mana-
patterns in the lives of people. When patterns gerial or executive level are responsible for col-
of communication behavior are confirmed or lecting and analyzing data to help organizations
discovered, scholars develop useful theories of and employees make more effective and efficient
communication that speak to the regularity of decisions. But are these examples of communica-
communication (Bostrom, 2003). tion research? They could be. Some organizations
The research techniques and methods pre- conduct surveys or focus groups to discover the
sented in this book are used to study the com- degree of effectiveness of their internal commu-
munication behavior of humans and the nication practices. Media organizations regularly
communication artifacts that people leave behind. use surveys or focus groups to discover if infor-
Although some people think of social science mational, advertising, or promotional messages
research as objective research, communication are being received as intended.
scholars use both quantitative (more objective) You could become a consultant and conduct
and qualitative (more subjective) methods— proprietary research, research that is commis-
sometimes separately and sometimes in combina- sioned by an individual or organization for its
tion with one another. Both types of methods are own use. Organizations use consultants to evalu-
empirical, meaning that both methods are based ate their internal communication systems and
on observations or experiences of communica- operational effectiveness. Political figures also
tion. Both types are needed because it is unlikely commission proprietary research to discover
that quantitative or qualitative methods alone can how they are doing in the polls and which of their
provide complete answers to the many questions central messages are having the most impact
we have about communication behavior. on potential voters. Marketing and advertis-
ing research is almost always proprietary. Even
Your Relationship with Research though the results of proprietary research are
private and intended only for the use of whoever
As discussed earlier, your relationship to this pays for the research, the researcher uses the
material can be conceptualized in two ways—as same procedures and practices used in conduct-
that of a researcher or as that of a consumer of ing scholarly or academic research.
research. You may take on the researcher role as Your relationship with research can also be
a student, as an employee, or as a consultant. It conceptualized as that of a consumer. You con-
is likely that the class for which you are reading sume the research of others when you read
this book will develop and conduct a research scholarly books and journals. You also consume
project as part of a class assignment. You may research when you see or hear personally or pro-
also decide that the process of research is inter- fessionally interesting information presented in
esting enough that you plan to continue your the media, and use information about goods and
education in graduate school, where you will services marketed to you. You might trust some
receive additional instruction in research meth- sources more than others—or be more cautious—
odology. You might even decide to become a if you knew how the data were collected and
professor and spend much of your professional analyzed.
time as a researcher, finding answers to ques- Right now, your role as a consumer of research
tions that interest you and matter to others. is more immediate than your current or potential
After you graduate, you might find your- role as a researcher. Your status as student forces
self in a professional position where research you into the consumer role as you collect infor-
is part of your regularly assigned job responsi- mation in the library or from the World Wide
bilities. Positions in marketing and advertising, Web to complete class assignments.

key06767_ch01_001-017.indd 3 1/25/10 7:15:16 PM

 Rev.Confirming Pages

4 CHAPTER 1 / INTRODUCTION TO COMMUNICATION RESEARCH

AN ETHICAL Is Communication Public or Private?

ISSUE
In general, what ethical issues do you believe are raised when researchers study
the communication behavior of others? About what communication situations
would you feel comfortable answering questions? In what situations would you
feel comfortable having a researcher observe you? Should some communication
contexts remain the private domain of participants, closed to researchers’ inquiries?
What about intimate communication between significant others in the privacy of
their bedroom? What about the communication of parent to child when discipline
is required? What about communication that occurs among co-workers as they
discuss ways to ridicule their boss? How would you respond if a communication
researcher asked you questions about your communication behavior during these
events? What arguments could you develop both for and against communication
scholars conducting research about such events? Should some communication
behaviors or contexts be off limits to communication researchers?

Your ability to evaluate the information you as an influence on your life and in your decision
collect has a direct impact on your ability to learn making. Throughout the rest of this chapter and
and prepare assignments. Moreover, the media throughout the book as well, specific examples
bombard us daily with information that has been of communication research will be highlighted
accumulated through research. In both instances, as we explore how research is conducted—that
your relationship with research requires that you is, how research is planned and carried out and
assume the role of detective. As a researcher, you how data are collected, analyzed, and reported.
are the primary detective seeking answers to The goals of the book are to provide you with
questions, fitting pieces of the puzzle together, the basic skills of a researcher and to enhance
and interpreting what you find to make conclu- your ability to be a better critic of the research
sions and recommendations. As a consumer, you reported by others.
must still be a detective sorting through the data
others have provided. In this role you still need
to distinguish good information from bad, test SCHOLARLY RESEARCH
assumptions and conclusions drawn by others,
and analyze the extent to which the research pro- With this introduction to research in general, we
cess others used fits your needs and situation. In will turn our attention to the formal and system-
this case, your job as a detective is to determine atic method of scholarly research. Researchers,
if the information you are using is misleading or or scientists, who have been trained in research
misinterpreted from its original source. methods and procedures (generally as gradu-
It is easy to feel overwhelmed or intimidated ate students) conduct research. These scholars
by the particular vocabulary and traditions of formalize their questions into research ques-
research. But if you approach learning about tions or hypotheses, which provide the scope
research as a detective would approach finding and direction of the research project as well as
information, you are likely to discover that for- guide the researcher in selecting quantitative
mal research is an extension of the types of infor- or qualitative methods to answer the questions.
mal asking and answering of questions that you The questions or hypotheses direct what data the
have done all your life. After reading this chapter, researcher collects. After the data are collected,
you should be able to identify how research acts the researcher or research team analyzes the

key06767_ch01_001-017.indd 4 1/25/10 7:15:16 PM

 Rev.Confirming Pages

SCHOLARLY RESEARCH 5

data to draw conclusions about the hypotheses of this passage, you are also a consumer and
or answer the research questions. Essentially, could verify my interpretation of their work by
conducting research is a matter of making claims going to the original source.
based upon data (O’Keefe, 2004). Different types
of claims require different types of evidence, or Goals of Research
data, which may be quantitative data, qualitative
data, or both. Accumulating knowledge through research is a
But the process is not complete. Scholarly, or continuous process. One research study cannot
academic, research is also public and available answer all the questions about any one issue or
to consumers. However, the process of mak- topic. This facet of learning—building on the
ing it public is certainly different than it is for research of others—is central to any academic
research conducted by a polling organization, discipline. Thus, the primary goal of commu-
for instance. Scholarly researchers describe what nication research is to describe communication
they have done in a paper that is submitted to phenomena as well as discover and explain the
a conference for presentation or to a journal or relationships among them. Continuing with the
book for publication. Other experts in the field example just given, discovery occurred when
review the paper. This review serves as a test. Kuhn and Nelson conducted research using
Have the authors used the best methodology to both quantitative and qualitative methodolo-
answer their questions or hypotheses? Have the gies to uncover how employees identified with
authors explained the results thoroughly and their organization, and how those identifications
logically? Are there critical flaws in the research resulted in communication networks. These
process that jeopardize the results? The papers scholars first built a case for their research ques-
that make it through the review process are then tion by drawing on the published research of
presented at a conference or published in an aca- other scholars. With the data they collected, they
demic journal or book. This is where the results were able to describe employees’ locations in
become consumable. their communication networks and their degree
Pick up a text that is assigned reading for of organizational identification. Finally, they pro-
one of your other communication courses. You vided an explanation of the relationship between
will find many references to research within the employees’ organizational identification and
chapters. As an example, the following passage their location in communication networks that
is from my text Communication and Organizational followed logically from their arguments and the
Culture: A Key to Understanding Work Experiences data. Thus, to put it more formally, research is
(Keyton, 2005): the process of discovery and explanation.
The research process, if approached system-
From any one position in the organization, it
atically, can have one of four results: It allows
may look like the culture is consistently singu-
the researcher to describe behavior, determine
lar. However, it is more typical for organiza-
causes of behavior, predict behavior, or explain
tions to structure themselves into networks
behavior. Describing behavior entails describing
based on tasks, relationships, information, and
outcomes, processes, or ways in which vari-
functions with organizational members iden-
ables (another name for the concepts we study)
tifying with, and belonging to, more than one
are related to one another. The following exam-
network (Kuhn & Nelson, 2002).
ple illustrates a research project that enabled a
The reference to the authors Kuhn and Nelson is researcher to describe behavior.
called an in-text citation. If you turned to the ref- Despite the number of contexts in which
erences listed at the back of the text, you would social support is given and received, detailed
find the publication information so you could descriptions of the ways in which social sup-
look up the 2002 journal article written by Kuhn port is enacted were missing. Using conversa-
and Nelson. As the author of the text, I relied on tions from a peer-to-peer telephone counseling
the research of Kuhn and Nelson. As the reader service, Pudlinski (2003) was able to develop

key06767_ch01_001-017.indd 5 1/25/10 7:15:16 PM

 Rev.Confirming Pages

6 CHAPTER 1 / INTRODUCTION TO COMMUNICATION RESEARCH

three categories to describe the ways in which and alternatively, employees who believed they
social support is constructed. Across a dataset were being fairly treated by their supervisors
of 366 responses from 44 calls, he discovered perceived that the organization’s policy was
three common patterns in the ways support pro- being upheld.
viders displayed social support: summarizing If researchers can describe communication
or minimizing the caller’s problem (e.g., “I see events and identify their causes, then they can
what you’re saying”), providing or seeking solu- turn to predicting behavior. If behaviors are pre-
tions (e.g., “What kind of a job do you think dictable, then we can anticipate what will hap-
you’ll apply for?”), and supporting a caller’s pen in the future. In turn, this knowledge can
report (e.g., “That’s good”). Pudlinski’s detailed help us make better decisions. For example, Sch-
descriptions of social support suggest that these neider, Lang, Shin, and Bradley (2004) wanted to
messages comprise greater variety and complex- determine if adding a story or narrative to jus-
ity than generally thought. tify game players’ actions in violent video games
Determining the cause or causes of behavior is would influence their game-playing experiences.
of interest to communication scholars because This research team wondered if the addition of
knowing the cause of something allows scholars a story-based justification of killing to save the
to later plan interventions or develop training world would influence game players differently
to increase the effectiveness of communication. than the weaker narrative structure found in vio-
Keyton, Ferguson, and Rhodes (2001) wanted to lent video games in which the story unfolds by
determine which variables had greater effect on moving to the next level of the game. Based on
employees’ perceptions that their organization the research literature, the research team hypoth-
lived up to its stated zero tolerance sexual harass- esized that video game players would iden-
ment policy. Would constructs in the interaction tify with characters and their goals to a greater
environment (e.g., respondent sex, being a target extent when a story is present than when a story
of sexual harassment, being treated fairly by co- is not present. Experienced video game players
workers and supervisors) influence how respon- participated in an experiment designed to test
dents characterized the sexual nature of the this prediction. Findings demonstrated that, yes,
organization’s environment and influence their video game players more strongly identified
perceptions of how well the organization lived with characters and their goals when a story was
up to its policy? Or would organizational rem- present. Thus, the research team’s prediction was
edies, such as receiving sexual harassment train- supported, suggesting that violent video games
ing and retaining knowledge from that training, with stories justifying violent acts may desensi-
be more influential? Using the responses of tize game players to violent behavior.
252 employees, the research team tested the Going beyond describing, determining causes,
influence of the interaction environment factors and predicting, explaining behavior means under-
against the set of organizational remedy factors. standing why a behavior occurs. For example,
Contrary to the presumption of the organiza- if researchers were able to determine how and
tion’s executives, organizational remedies were why health campaigns work, more effective
not related to employees’ perceptions about the campaigns would ultimately result in a healthier
organization’s zero tolerance sexual harassment society that spends less money on health care.
policy. The study’s findings did support, how- But finding such an explanation is difficult and
ever, that factors in the interaction environment often requires a series of sophisticated research
caused two different employee reactions. First, projects. Working from a well-developed and
male employees, employees who had been vic- validated theoretical basis is another way to
tims of sexual harassment, and employees who develop explanations for communication behav-
believed they were not being fairly treated by ior. For example, A. J. Roberto, Meyer, Boster,
their co-workers perceived the organization’s and H. L. Roberto (2003) surveyed 488 junior
environment as sexualized and that the organi- high students about four aggressive behaviors:
zation did not live up to its stated policy. Second, watching a fight, telling friends about a fight that

key06767_ch01_001-017.indd 6 1/25/10 7:15:16 PM

 Rev.Confirming Pages

COMMUNICATION AS A SOCIAL SCIENCE 7

is going to happen, insulting others, and fight- Theoretically driven research is built on the
ing. For each of the aggressive behaviors except results of previous researchers, and it provides
fighting, the explanatory model provided by the a foundation for subsequent researchers. Theory
theory of reasoned action (i.e., the best determi- cannot be formulated, tested, and verified in
nant of actual behavior is behavioral intention) one research study. Rather, theory is developed
explained students’ participation in aggressive and tested over time. What we come to know as
behaviors. That is: students’ attitudes about a the theory to explain some phenomenon is the
behavior created behavioral intention, which, in result of many research studies and the efforts of
turn, caused their participation in that behavior. many researchers.
These four outcomes—description, determina- Cushman (1998, p. 9) points out that “human
tion of causes, prediction, and explanation—are communication is one of the most creative, flex-
closely related. New knowledge in one area will ible, and thus anti-theoretic processes in which
affect how questions are asked and answered in human beings engage.” Why? The complex-
another. ity of communicating in multiple cultures with
multiple, and sometimes conflicting, social goals
Research and Theory provides the opportunity for multiple individ-
ual interpretations. Moreover, communication
When researchers discover that one explana- occurs in multiple languages with different sets
tion about the relationship between phenomena of rules and practices. According to Cushman,
occurs regularly, a theory can be constructed. this variability is one important reason commu-
Although many definitions exist for the term nication scholars must look for the mechanisms
theory, in general, a theory is a related set of or constructs that are constant regardless of the
ideas that explains how or why something hap- language used to communicate. Thus, commu-
pens. In other words, a theory provides a way nication researchers use systematic procedures
for thinking about and seeing the world (Deetz, and scientific principles to conduct research
1992). More formally, a theory is a set of inter- about how and why humans communicate as
related concepts, definitions, and propositions they do.
that presents a systematic view of phenomena.
A theory specifies the relationships among the
concepts with the objective of explaining and pre- COMMUNICATION AS
dicting the phenomena being studied (Kerlinger, A SOCIAL SCIENCE
1986). As a result, theory helps us understand or
make sense of the world around us. Of course, There are many methods of discovery and expla-
communication theories can help us understand nation, or many ways to view communication
our own communication behaviors as well as the problems. Scholars conduct their research from
communication behaviors of others (Miller & paradigms that provide different explanations
Nicholson, 1976). and functions for the role of symbols, messages,
With respect to communication, a theory is and meanings in the process of communication.
one or more propositions about people’s commu- These paradigms also create differences in what
nication behavior that enables a communicator to researchers count as data. You have probably
figure out how to communicate with particular explored these different paradigms in courses on
individuals or in a given situation. The term the- communication and rhetorical theory.
ory, however, does not have one precise meaning. Broadly, this book explores the social sci-
Rather, different definitions of the term are used entific study of communication for which a
because they promote different approaches to wide variety of methods is available. This text
research (Craig, 1999; Miller & Nicholson, 1976). will introduce you to both quantitative meth-
The best research is driven by theory, validates a ods (generally speaking, research that relies on
theory, further explains a theory, challenges an numerical measurement) and qualitative meth-
existing theory, or aids in the creation of theory. ods (generally speaking, research in which the

key06767_ch01_001-017.indd 7 1/25/10 7:15:16 PM

 Rev.Confirming Pages

8 CHAPTER 1 / INTRODUCTION TO COMMUNICATION RESEARCH

researcher is the primary observer or data collec- and oppression that can occur through commu-
tor). Both methods are part of the social science nication practices and structures. (For example,
research tradition as practiced in the communi- what ideological structures in our society control
cation discipline and reported in communication or dominate the dissemination of new media
and related-discipline journals and scholarly technology?) Some critical scholars use qualita-
books. Both quantitative and qualitative meth- tive methods in their research, and examples are
ods of research are empirical; that is, both meth- included in this book. The definitional boundar-
odologies are based on or are derived from ies for what constitutes these three perspectives
experiences with observable phenomena. This is for studying communication (social science,
the critical element of research. Both quantitative rhetorical, critical) are blurry, and not mutually
and qualitative methodologies can observe and exclusive. Critical communication research can
describe human communication. And both can also be rhetorical. But, broadly speaking, this
help researchers in explaining or interpreting text focuses on the social scientific methods for
what was observed. conducting communication research.
The study of communication from a social How does the study of communication differ
science perspective uses quantitative or qualita- from the study of other social sciences? Gener-
tive methods to look for patterns of messages or ally, the social sciences are defined as those areas
communication behaviors. These patterns can be of scientific exploration that focus on the study
based on observations or measurements across of human behavior. Psychology, sociology, and
the experiences of many individuals or on the political science are other fields in the social sci-
in-depth observations from one case over time. ences. As a social scientist, the communication
Either way, the data must be empirical; that is, scholar focuses on symbols used to construct
the data must be able to be verified through messages, messages, the effects of messages, and
observations or experiences. their meanings.
How does the study of communication as a The social sciences are different from the nat-
social science differ from humanistic and critical ural sciences in that the social scientists focus on
studies of communication? The study of com- the study of human behavior. Problems that are
munication from a rhetorical perspective often significant for study in the social sciences involve
focuses on how language is used to persuade in several important variables, and untangling the
a particular case (for example, a specific speech effects of one variable from another is difficult.
by a specific person or other one-time event from Moreover, the social sciences recognize that the
which a text can be drawn or developed). In researcher is a human instrument with biases
addition to the rhetorical event itself, an analysis and subjective interpretations that can affect the
would include the historical, cultural, and social individuals or processes under investigation.
contexts surrounding it. Probably the most use- Finally, seldom can an entire system of human
ful distinction is that rhetoric is planned for a behavior (for example, an entire organizational
specific goal for a specific audience, whereas the communication system) be observed. Even if it
social science study of communication focuses could be, human systems are always subject to
on the interactive moment between and among new influences; thus, what is being observed
conversational participants. A rhetorical study is is dynamic. As a result of these differences, the
more focused on one case, whereas the social sci- study of human behavior is difficult to isolate
ence study of communication looks for patterns and control even if the examination is done in
across people or situations. the laboratory setting.
From a critical perspective, the research One last point is that social science research
emphasis is on the hidden assumptions of is contextually and culturally bound. Research
broad social structures that serve the interests is contextualized first by the number and type
of some people (those in power) more than oth- of people participating and by the type of com-
ers. Critical communication scholarship focuses munication being investigated. Second, research
on understanding the domination, inequality, is contextualized by where the investigation

key06767_ch01_001-017.indd 8 1/25/10 7:15:16 PM

 Rev.Confirming Pages

COMMUNICATION AS A SOCIAL SCIENCE 9

occurs—in the lab or in the field. Third, research If the researcher cannot formulate a tenta-
is contextualized by the culture in which it occurs. tive proposition after reviewing the existing lit-
Researchers and participants bring cultural erature, then a research question is developed.
norms and values to what they do and how they A research question asks what the tentative
communicate. All these contextual and cultural relationship among variables might be or asks
factors influence the research investigation, the about the state or nature of some communica-
data produced, and the interpretation of results. tion phenomenon. For example, we used the
research question “Will there be a relationship
The Scientific Approach between ethical ideology and the ability to accu-
rately distinguish between verbal and nonverbal
So how do communication researchers incor- behaviors that have been shown to be associated
porate scientific characteristics into the process with flirting and sexual harassment?” (Keyton &
of conducting research? Generally, research Rhodes, 1997, p. 135). Although numerous stud-
follows procedural traditions that have been ies had been published on both ethical ideology
tested, validated, confirmed, and accepted by and sexual harassment, no study had explored
social scientists of many disciplines over time. the relationship between these two issues. Thus,
The research process has five general steps (Ker- we posed a question to help us determine if a
linger, 1986). Figure 1.1 illustrates this process. relationship occurred. We could not propose
First, researchers start with a question that what type of relationship would exist.
interests them. A question may arise from their In the third step, which is often underempha-
personal experiences or from experiences that sized, the researcher uses reason and experience
have been reported to them by others. Or a ques- to think through the hypotheses or research ques-
tion may arise from reading the scholarly or con- tions that are developed. A researcher might ask,
sumer literature. In other words, some question, “Do the research questions and hypotheses I’ve
or curiosity, is not explained or inadequately generated capture the essence of the problem?” or
explained. A question may also be stated as a “Are there other variables that affect the relation-
problem. In either form, the researcher cannot ship between the two variables I’ve identified?”
continue the research process without identify- This step of reasoning, or thinking through,
ing and specifying the question or problem. For may, in fact, change the research agenda. It
example, my own curiosity about why sexual may broaden the nature and scope of research,
harassment continues to occur in organizations or it may more narrowly focus the researcher’s
despite clear societal and organizational signals inquiry. By taking this step in refining and for-
that a perpetrator faces employment, legal, and mulating the research question or hypothesis,
even financial consequences for sexually harass- researchers discover the most significant issue
ing another employee caused me to pursue this that can be addressed given their initial ques-
area as a topic of research. tions or problems. By using the experience we
Second, the researcher uses the question or gained in developing sexual harassment training
problem to formulate a hypothesis, or a tenta- for organizations and by searching the literature,
tive, educated guess or proposition about the we discovered that one of our proposed hypoth-
relationship between two or more variables. eses (“participants who identified themselves
Oftentimes, hypotheses take the form of state- as targets of sexual harassment would iden-
ments such as, “If x occurs, then y will follow” tify more verbal and nonverbal cues as harass-
or “As x increases, so will y.” With respect to our ment”) would not adequately explain why some
sexual harassment research, we used previous employees view behaviors as sexual harassment
scholarship to help direct our inquiry. One of and others do not. In other words, an employ-
our hypotheses proposed that participants who ee’s perceptions of sexual harassment would not
identified themselves as targets of sexual harass- simply turn on whether she or he had been sexu-
ment would identify more verbal and nonverbal ally harassed. As a result, we tested three other
cues as harassment (Keyton & Rhodes, 1999). explanations.

key06767_ch01_001-017.indd 9 1/25/10 7:15:16 PM

 Rev.Confirming Pages

10 CHAPTER 1 / INTRODUCTION TO COMMUNICATION RESEARCH

Symbols
Meanings
Messages

Communication
knowledge

Reason
and
experience

New
knowledge

Design and conduct study

FIGURE 1.1 General Steps of the Scientific Approach.

Fourth, the researcher designs and conducts data are analyzed and interpreted in reference to
the observation, measurement, or experiment. the question or hypothesis posed in step 2 and
Although each variable or element identified refined in step 3.
in the research question or hypothesis must be Thus, the social scientific approach to com-
observed or measured, it is actually the relation- munication research starts with a problem, a
ship between them that is assessed. Fifth, the question, or an idea as the researcher identifies a

key06767_ch01_001-017.indd 10 1/25/10 7:15:16 PM

 Rev.Confirming Pages

COMMUNICATION AS A SOCIAL SCIENCE 11

barrier or gap in knowledge. Then, the research to be false, then, logically, it cannot be true, or
question or hypothesis is formulated. Once valid. If the proposition and its explanation hold
developed, the research question or hypothesis up over time, scientists come to accept the find-
is revisited and refined. Only then can the meth- ing as true or real, until shown otherwise.
odology be designed and carried out. The results 4. The results of a research study are replicable,
are interpreted and fed back into our knowledge or repeatable. Ideally, different researchers in dif-
of the original problem. As a result, the problem ferent settings and with different participants
is resolved, completely or partially, or new ques- should conduct replication studies—studies that
tions arise. Recognize that the five steps described repeat the same procedures. The results of any
are not necessarily discrete. One step blends one study could be wrong for many reasons.
into another. Work in one step may require the Repeating the same or a very similar study many
researcher to go back and revise what was previ- times and obtaining the same or very similar
ously completed. results ensures that the finding is real and can be
counted on.
Characteristics of Science 5. For replication to occur, research must be part of
In pursuing these five steps of the research pro- the public record. This is why communication schol-
cess, researchers can select from a variety of ars publish their work in academic journals and
quantitative and qualitative methods. Although scholarly books. Scholars typically are not paid
individual methods vary in the extent to which for these publications, but their work is supported
they encompass the following 12 characteris- through their universities and sometimes by gov-
tics, over time these characteristics have distin- ernment agencies and other funding organiza-
guished scholarly research from everyday, or tions. As part of the public record, university and
informal, ways of knowing (Andersen, 1989; college libraries provide access to these journals
Bostrom, 2003; Katzer, Cook, & Crouch, 1978; and books so you can scrutinize what researchers
Kerlinger, 1986). These characteristics are not did and how they did it. Scientific study is avail-
unique to the study of communication. Rather, able to other researchers and the general public. It
scientists of all disciplines have accepted them. is not private, or proprietary, research conducted
Thus, the tradition of science rests with these 12 for the exclusive use of those who paid for the
characteristics: research to be done. Because scientific research
is part of the public record, scholars build onto
1. Scientific research must be based on evidence. as well as challenge each other’s work. All pub-
Even experts can disagree. That is why evidence, lished research includes a section describing the
or data, is paramount to the research process. methods by which the data were collected and
Further, scientific research is based on the prin- interpreted. This allows others to evaluate the
ciple of empiricism. This means that careful methods used for potential weaknesses and to
and systematic observation must occur. What is replicate the study for further validation.
observed and measured—the data—serves as the 6. Because scientific research is part of the public
evidence researchers use in making their claims. record, it is also self-correcting. This characteristic
2. Scientific research is testable. This means that means that the scholars who conducted the orig-
the proposition, research question, or hypothesis inal study as well as the scholars who replicate
must be able to be probed or investigated with or challenge studies are continually improving
some quantitative or qualitative methodology. the methods by which they observe or measure
If the proposition cannot be tested or challenged the phenomenon of interest. Improving on the
in a research study, only speculations about the methods is one way to develop a greater under-
validity of the claim can be made. standing and more detailed explanations.
3. Researchers must explore all possible explana- 7. Scientific research relies on measurement and
tions in an effort to demonstrate that their proposition observation. Researchers can measure your com-
cannot be disproved. If a proposition can be shown munication apprehension, for example, by

key06767_ch01_001-017.indd 11 1/25/10 7:15:17 PM

 Rev.Confirming Pages

12 CHAPTER 1 / INTRODUCTION TO COMMUNICATION RESEARCH

asking you to fill out a questionnaire. Or they intimacy of the researcher–participant relation-
can observe your apprehension by counting the ship, scholars doing this type of research must
number of times you lose your place when you be able to describe their role in the research
are speaking and have to refer to your notes. process, and this act requires a certain amount
When something is not directly observable, of objectivity. Alternatively, statistics must be
researchers develop and rely upon other meth- selected and statistical findings must be inter-
ods (such as questionnaires) to capture partici- preted—both subjective decisions. The point
pants’ attitudes, perceptions, and beliefs. here is not to quibble over the alignment of
8. Scientific research recognizes the possibility objectivity/subjectivity to quantitative/quali-
of error and attempts to control it. When things tative method. Rather, it is to introduce the con-
are measured or observed, we expect that some cept of scientific objectivity as practiced by all
error will occur. For example, errors occur when researchers regardless of which methodology
a researcher does not see the participant lose her they choose.
place while speaking because his attention is dis- 10. Science by its nature rests on an attitude of
tracted by loud music playing in another room skepticism. By their nature, researchers are sus-
or when a mistake is made in transferring data picious; they do not rely on what appears to be
from the coding sheet to the spreadsheet. Errors obvious or on common sense. Within the social
can occur in many places in the research process. science research tradition, researchers rely on
Quantitative research limits and accounts for data compiled from quantitative and qualitative
error through the use of systematic procedures methodologies to answer their questions and
and statistics. Qualitative research accounts for support their claims. This element of skepticism
error by providing detailed description to allow is what allows, even encourages, researchers to
the reader to draw his or her own conclusions put their assumptions through a process of test-
and interpretations. Most procedures have been ing or verification.
standardized over time and across disciplines. 11. Scientific research has an interest in the gener-
Such formality in procedure acts as a form of alizability of findings, or the extension of the findings
control to help the researcher eliminate error, to similar situations or to similar people. In quan-
bias, and other explanations for the result found. titative research, findings are more externally
Despite these control mechanisms, it is impos- valid if they apply to a range of cases, people,
sible to eliminate all bias and error in conduct- places, or times. In other words, are the results
ing research. Recognizing that bias and error can of studies that use traditional college-age stu-
occur, researchers must take every precaution in dents as research participants applicable to non-
helping to minimize it. traditional college-age students? What about
9. Scientific objectivity requires the researcher to teenagers? Or retired adults? All studies have
minimize personal bias and distortion. Despite the limitations, but by using discipline-accepted
passion for their topic and the time devoted to procedures, researchers can help strengthen the
the project, researchers cannot be so commit- generalizability of their results. In qualitative
ted to their own point of view and expectations research, findings are typically less generaliz-
that they fail to see other explanations when able because they are more case-specific. How-
they appear. In essence, the objectivity of sci- ever, the generalizability of qualitative results
ence distinguishes it from conclusions based can also be strengthened as a researcher spends
solely on opinion. Too frequently, objectivity is greater lengths of time observing research
associated only with quantitative research, and participants.
subjectivity is associated only with qualitative 12. The final characteristic of science is its heuris-
research. In reality, all researchers, regardless tic nature. This means that research findings lead
of method, must demonstrate objectivity in to more questions. At the conclusion of most
conducting research. Even though qualitative journal articles, scholars identify new questions
research is more subjective due to the greater that surface from their findings. The ability of a

key06767_ch01_001-017.indd 12 1/25/10 7:15:17 PM

 Rev.Confirming Pages

WHAT KINDS OF QUESTIONS DO COMMUNICATION SCHOLARS ASK? 13

finding to suggest additional questions or new answer all questions. Although you will find that
methods of conducting the research is its heu- you are drawn to some methods more naturally
ristic ability. The ultimate objective of science than others, you will develop stronger analyti-
should be to lead scientists to future discoveries cal skills, both as a researcher and as a consumer
and investigations. of research, if you develop skills collecting and
interpreting data from a variety of methodologi-
Methodological Extremes cal techniques.

This introductory chapter is a good place to

also introduce you to a methodological extreme WHAT KINDS OF QUESTIONS DO
that you should be aware of as you learn about COMMUNICATION SCHOLARS ASK?
research methodology (Bouma & Atkinson,
1995). A child given a hammer for the first time Among the variety of questions that can be asked
is likely to run around the house and ham- about communication are both important ques-
mer anything and everything. From the child’s tions and trivial ones (Miller & Nicholson, 1976).
viewpoint hammering is fun, even if hammer- How do researchers determine the significance
ing on Mother’s new table is neither desirable of a question? There are three criteria: personal
nor appropriate (at least from the mother’s interest, social importance, and theoretical sig-
point of view). The child hammers because it is nificance. “Any question that interests or per-
new and novel. plexes a person is personally significant” (Miller
Unfortunately, this same phenomenon can & Nicholson, p. 49). Questions that interest me
exist when students begin to learn research include (1) How do children learn to communi-
methods (see Cappella, 1977; Janesick, 1994). cate in groups? (2) Why do some employees per-
With each new technique, there is the ten- sist in sexually harassing other employees given
dency to believe that this particular method the individual, relational, professional, financial,
can answer any question. However, think of and legal consequences that are likely to result?
the method (or hammer) as a tool and recog- (3) To what extent does the relational develop-
nize that there are appropriate tools for differ- ment among group members affect the task
ent purposes. To expand the tool metaphor, effectiveness of a group? Some of the studies my
hammers are good for pounding in nails, but colleagues and I have conducted in these areas
screwdrivers are better for twisting in screws. are used as examples in this book.
Saws are good for cutting things in two but are What questions interest you? They are prob-
ineffective for affixing something with a nail or ably related to your personal or professional
a screw! The point here is that the substantive life. Your interests may be idiosyncratic, not
content of the research question or hypothesis coinciding with the interests of others. This
drives the selection of the methodological tool demonstrates why the second criterion of social
(Hackman, 1992; Janesick, 2000). importance is valuable and necessary. Because
Methods are useful or effective only to the communication is a social activity, significant
degree that they help the researcher answer a spe- questions are those that have a general social
cific question or explore a specific hypothesis. In import (Miller & Nicholson, 1976). For exam-
fact, the answers produced through any method ple, “What media campaigns would decrease
of investigation are influenced by the specific the likelihood that children try drugs and alco-
methodological technique used in the investiga- hol?” or “What negotiation strategies work best
tion (Clark, 1990). So, if you let the method drive in resolving international differences?” Finding
the research questions you ask or the hypotheses the answers to questions like these could have a
you test, then your results are more likely to be powerful impact on many lives. Questions that
tied to the method you selected than to repre- drive research do not have to relate to all mem-
sent a valid response to the question or test of bers of society. But we should ask who would
the hypothesis. No one research method can be affected by the answer. If enough people are

key06767_ch01_001-017.indd 13 1/25/10 7:15:17 PM

 Rev.Confirming Pages

14 CHAPTER 1 / INTRODUCTION TO COMMUNICATION RESEARCH

affected by or could use the answer to the ques- procedures shaped their use of that medium.
tion, then the question has social importance. The descriptions provided by both studies move
The third criterion is theoretical significance. the phenomena from the abstract realm into the
Questions that further the construction of com- specific. Without defining phenomena of inter-
munication theories are significant (Miller & est, it would be impossible to ask other types of
Nicholson, 1976) because they deepen our under- questions.
standing and explanation of communication After the what has been adequately defined,
behavior. When these questions are posed and researchers generally turn to questions of rela-
answered by research, we gain new knowledge. tionships. These questions of variable relations
Keeping these three criteria in mind can help examine if, how, and the degree to which phe-
us respond to the “So what?” question. Many nomena are related. For example, Metzger and
times, people read research reports and have Flanagin (2002) asked how age, education, sex,
difficulty finding any significance or utility for and Internet experience related to a relative
the findings. If your research project is driven instrumental orientation for using media. The
by personal interest, has societal significance, researchers measured college students’ instru-
and helps further theory—and if these issues are mental (i.e., intentional, selective, goal-driven)
described in the research report—then the “So and ritualized (i.e., habitual, passing time) use
what?” has been answered. of four new media. To create the relative score,
the mean scores for instrumental and ritualized
The Nature of the Questions technology use were compared. This score was
then assessed to see if it differed according to
As you read the communication research lit- student ages, education levels, sex, and Inter-
erature, you will notice four types of questions net experience. By doing so, the researchers
(Stacks & Salwen, 2009). The first type, questions could describe the relationship between rela-
of definition, provides definitions for phenom- tive instrumental orientation and the four other
ena in which we are interested. Whereas you variables. In this case, only age was related to
may believe that all definitional issues have been relative instrumental orientation. Older users
addressed, remember that new communica- used new media more instrumentally than did
tion situations and environments and changing younger users.
societal values create new areas to explore and By understanding how variables are related,
define. As a question of definition, Olson and we have a greater understanding of our world
Golish (2002) asked what topics of conflict are and the role of communication behavior in it.
associated with the use of aggression in roman- Most important, questions of variable relations
tic relationships. In this study, the researchers help the community of communication scholars
wanted to discover the topics aggressive couples build and develop theory.
argue about. Some individuals reported as few Questions of value, the third type, ask for
as one topic; others reported as many as seven individuals’ subjective evaluations of issues and
topics of argument in their romantic relation- phenomena. Questions of value examine the aes-
ships. Analysis of these incidents resulted in thetic or normative features of communication,
nine categories of argument. Another example asking, for example, how good, right, or appro-
of communication research that asks questions priate a communication phenomenon or practice
of fact is a study that examined how individuals is. Questions of value are inherent in a study that
used e-mail to communicate in romantic relation- explores how everyday discourse stigmatizes
ships at work (Hovick, Meyers, & Timmerman, teenagers who are homeless. Harter, Berquist,
2003). Participants in the study responded to Titsworth, Novak, and Brokaw (2005) inter-
an online survey allowing the research team to viewed homeless teens, educators, and social
answer how frequently partners in workplace service providers. These teens, often called the
romances used e-mail to communicate at work, hidden homeless, try to disguise the fact that they
and to what extent organizational policies and are homeless when talking with others to avoid

key06767_ch01_001-017.indd 14 1/25/10 7:15:17 PM

 Rev.Confirming Pages

WHAT KINDS OF QUESTIONS DO COMMUNICATION SCHOLARS ASK? 15

TRY THIS! Evaluating Communication Questions

For each of the questions listed, evaluate your personal interest in the question and
the question’s social importance. Use the table to capture your evaluations. Rate
your personal interest on a scale of 1 to 5, with 1 being “little or no personal inter-
est” and 5 being “high personal interest.” Rate social importance on a scale of 1 to 5,
with 1 being “little or no social importance” and 5 being “high social importance.”
Personal Social
Preliminary Research Question Interest Importance
What is the prevalence and context of verbal aggression
in televised professional wrestling? (From Tamborini,
Chory, Lachlan, Westerman, & Skalski, 2008)
Will news stories that are more deviant (i.e., the story
is about an event that is unusual, breaks some norms,
may lead to some kind of change in society) be more
likely to be selected? (From Lee, 2008)
Will fund-raising messages that contain information
about fund-raising goals result in higher donations?
(From Das, Kerkhof, & Kuiper, 2008)
Will employees who are more centrally located in
their friendship network at work be less likely to quit?
(From Feeley, Hwang, & Barnett, 2008)
What tactics do employees report using when expressing
disagreement or contradictory opinions to their
supervisors and managers? (From Kassing, 2009b)
What coping strategies do people use when dealing
with unfulfilled standards, or preexisting beliefs,
about what makes a good dating relationship? (From
Alexander, 2008)

Compare your evaluations with those of other students. How are your evalua-
tions similar or different? What other questions about communication do you
believe merit researchers’ attention?

being stigmatized or labeled. Other interviews Questions of policy are the fourth type. Com-
revealed that community members are gener- munication researchers seldom test policy issues
ally unaware of this homeless population and directly, but the results of research studies are
the difficulties the teens encounter trying to con- often used to recommend a course of action.
tinue their education. This study raises the ques- Roberto, Carlyle, Zimmerman, Abner, Cupp,
tion of how this type of public discourse inhibits and Hansen (2008) tested a 7-week interven-
conversations that could bring awareness to the tion program designed to prevent pregnancy,
problem and help the teens and their families. STDs, and HIV in adolescents. The intervention

key06767_ch01_001-017.indd 15 1/25/10 7:15:17 PM

 Rev.Confirming Pages

16 CHAPTER 1 / INTRODUCTION TO COMMUNICATION RESEARCH

included six computer-based activities. Over 300 5. The best research is that which is driven by
10th graders at two high schools participated theory, validates a theory, further explains
in the study. Students at one high school com- a theory, challenges an existing theory, or
pleted the intervention activities; students at aids in the creation of theory.
the other high school did not and served as the 6. As a social science, communication
control group. Students who participated in the researchers use both quantitative and quali-
intervention program outperformed students in tative methods.
the control group on disease knowledge, con-
7. The study of communication from a social
dom effectiveness, how to negotiate condom
science perspective looks for patterns across
use, and attitudes toward waiting to have sex.
cases and focuses on symbols used to con-
The study demonstrated that modest computer-
struct messages, messages, the effects of
based interventions could be effective. Because
messages, and their meanings.
this type of intervention can be used to reach a
large number of teens, the findings have policy 8. Communication scholars start with an inter-
implications for agencies considering how to esting question and then formulate a formal
allocate funds for these types of health-related research question or hypothesis.
programs. 9. Questions suitable for communication
As you can see, communication research var- research are those for which the researcher
ies widely in its subject matter. Some research has a personal interest, one that is of social
has implications for the development of com- importance, and one that has or can help
munication theory, some has more practical develop theoretical significance.
application, and some contributes to both theory 10. A hypothesis is a tentative, educated guess
and practice. But all research starts with a basic or proposition about the relationship
question about communication that needs an between two or more variables.
answer, and all research uses some form of sci-
entific and systematic research methodology in 11. A formal research question asks what the
providing those answers. tentative relationship among variables
might be, or asks about the state or nature
of some communication phenomenon.
SUMMARY 12. Research is judged to be scientific by
12 characteristics: its empirical nature, its
1. Research is asking questions and finding ability to be tested, the extent to which it
answers. can be falsified or disproved, the ability
to replicate or repeat findings, the pub-
2. Scholarly research is the discovery lic nature of findings, its self-correcting
of answers to questions through the nature, the ability to measure or observe
application of scientific and systematic the phenomenon of interest, the ability
procedures. to minimize error through the control of
3. Academic research follows accepted norms procedures, its level of objectivity, the
and procedures that have been adopted by skepticism it raises, the generalizability of
scholars from many disciplines. findings, and its heuristic nature.
4. In the process of scientific discovery and 13. Questions suitable for communication
explanation, four outcomes are sought: research may be questions of fact, questions
describing behavior, determining causes of of variable relations, questions of value, or
behavior, predicting behavior, and explain- questions of policy.
ing behavior.

key06767_ch01_001-017.indd 16 1/25/10 7:15:17 PM

 Rev.Confirming Pages

KEY TERMS 17

KEY TERMS See the website www.mhhe.com/keyton3 that

accompanies this chapter. The site contains
empirical questions of value a chapter summary, outline, and checklist. A
PowerPoint presentation created by Dr. Key-
heuristic questions of variable ton is also available for download as well as
hypothesis relations a short multiple-choice quiz to check your
proprietary research research learning. You will also find a list of Internet
qualitative methods research question resources that provide further information
social science research about the issues in this chapter.
quantitative methods
questions of definition theory
questions of policy

key06767_ch01_001-017.indd 17 1/25/10 7:15:17 PM

 Rev.Confirming Pages

CHAPTER TWO

The Research Process:

Getting Started

Chapter Checklist

After reading this chapter, you should be able to: 6. Glean the basic ideas from reading the
abstract, literature review, and discussion
1. Explain why the research process starts with
sections of a research article.
identifying a research problem.
7. Track a citation back to its original source.
2. Develop a preliminary question from a topic
or issue. 8. Effectively summarize and report what you
have found in the library.
3. Explain why a preliminary question is supe-
rior to a topic in conducting library research. 9. Describe what a theory is and its role in
communication research.
4. Evaluate preliminary questions for their
completeness and clarity.
5. Conduct a search for print and online
resources.

key06767_ch02_018-034.indd 18 1/30/10 7:38:35 AM

 Rev.Confirming Pages

THE RESEARCH PROCESS MODEL 19

Doing research means joining the conversa- because your ability to respond to the changing
tion. Whether you are conducting a literature environment presented to you is really the key
review for a class assignment or developing issue and determines whether you are success-
a literature review to support a research proj- ful in answering your question. Just as detectives
ect you design and conduct, you will need to conduct their investigations within the letter of
know what exists in the research literature. the law, researchers conduct their investigations
This chapter will help you identify ways of according to the traditions of scholarly research.
identifying research ideas and turning them If you were trying to explain how a detective
into preliminary questions. After reading conducts an investigation to someone who was
this chapter, you should be more comfortable unfamiliar with the process, you might, in gen-
and effective in conducting literature reviews eral, say that detectives seek answers to unan-
and in determining what is a credible source. swered questions, assess the situation presented
Much of what you will find in the library will to them, and then identify the procedures most
be related to the theories researchers use to likely to answer the questions presented by
describe, predict, and explain communication the situation. In the ideal situation, when the
behavior. This chapter will also describe what investigation was over, the questions would
theory is and how it is developed through the be answered. But detectives would not be able
research process. to explain in advance exactly how to find the
answer to any one question because they would
not be able to predict which clues they would
THE RESEARCH PROCESS MODEL uncover and which they would not. Certainly,
an experienced detective has developed strate-
In Chapter 1, I introduced the metaphor of gies that can help with the investigation. As a
researcher as detective. Like detectives, research- researcher, you can rely on scholarly standards,
ers are seeking answers to questions. There are traditions, and norms to help you answer your
two possibilities regarding the information they question. Yet, in neither case would your strate-
need. First, an answer may already exist, but gies or predictions be certain.
that information is not known to the researcher. Two models—or strategies—can provide a
In this case, library research usually provides general explanation of what the research pro-
the answer. Second, an answer is neither known cess is about, how one conducts research, and
nor available. In this case, the researcher must what one expects to be able to conclude at the
develop and conduct research to uncover end. Take a look at the first research model, pre-
an answer. In either case, finding an answer sented in Figure 2.1, to see how the deductive
depends on the researcher’s detective skills, research process is structured. Notice how the
or the ability to search and track down informa- model is circular and cyclical. Each of the steps
tion that fits his or her needs. must occur for the research process to be com-
The model for conducting research is similar plete. In this case, after identifying the research
to the plan a detective would follow in con- problem, the detective or researcher begins with
ducting an investigation. There are rules to be a theory and then gathers evidence, or data, to
followed and multiple paths that can be taken assess whether the theory is correct. This type
according to the questions asked. Once started, of research process is deductive because the
the research process or investigation proceeds researcher is moving from a known or assumed
logically and steadily. Working from what position supported by a theory to the particulars
you already know and understand, your pri- of the data.
mary objective as detective or researcher is to After entering the research process where it
find information that answers the questions. begins—“Identify the research problem”—the
Yet, obstacles and pitfalls along the way may researcher uses theory to guide the investigation.
keep you from accomplishing your goal. You Next, based upon theory and research findings,
must be vigilant and pay attention throughout the researcher formulates the research question

key06767_ch02_018-034.indd 19 1/30/10 7:38:35 AM

 Rev.Confirming Pages

20 CHAPTER 2 / THE RESEARCH PROCESS: GETTING STARTED

Deductive
research
process
starts here

Ask new Identify the

questions research problem R
re ev
se

ie rch
w fi
a
th ndi
eo ng
at a

ry
interp and
ret d

and
e
Analyz

s
th e esti o n
h y ch q t h e
sis
o r s e ar l a t e
u
Ga

re rmu

po
th e

Fo
rd

a
at

D e sig n t h e
m etho d s a n d
procedures

FIGURE 2.1 The Deductive Research Model

or hypothesis. Continuing on from there, the around the foundation of a research question
researcher then selects the research methods that (Figure 2.2). After the data are gathered and
will help in answering the questions or hypoth- examined, theories are developed in response
eses. Then data are gathered and interpreted. to what the data reveal. This type of research
Although the researcher will be able to answer process is inductive because the researcher
the initial questions at this point, the research is moving from the specifics of the data to the
process is not necessarily complete. more general explanation of theory. Again, the
Recall that research is prized for its heuristic research process is complete, but only temporar-
characteristic. If research has heuristic signifi- ily. Reports of these findings are likely to encour-
cance and values building on the work of oth- age researchers to identify new research topics
ers, answering one question should lead to other and start the process again.
questions for which answers are needed. Thus, Regardless of where one enters the research
as answers are developed from the interpreta- process, all of the steps are linked together. The
tion of data, the research process starts over steps are not independent activities but rather
again with a new question. are interdependent. At times, researchers believe
Alternatively, a detective suspends judg- they have completed a step and proceed to the
ment in beginning his or her detective work and next—only to find that they do not have the most
develops a plan for gathering data that is framed effective foundation from which to proceed.

key06767_ch02_018-034.indd 20 1/30/10 7:38:35 AM

 Rev.Confirming Pages

THE RESEARCH PROCESS MODEL 21

Inductive
research
process
starts here

y the
ntif Co
Ide earch problem res nsid
res ea er
Ask new rch
fi

ex ndin
questions

ist gs
ing
y
Develop theor

Develop
the research
question
A n erpr
int

re nd
du ds a e
aly et

o n th

s
ig
ze d

n es
D et h e
a

at d m roc
a
p
G ath
er d ata

FIGURE 2.2 The Inductive Research Model

And so they must go back and work through the inductive model, which typically relies on
the preceding step again. As you will discover qualitative methods, is described in greater
in the class for which you are reading this book, detail in Chapter 4. Throughout the research
research is not evaluated solely on its outcomes. process, researchers must be concerned with
Rather, the process that leads to the outcome, or issues of ethics and integrity. Communication
research result, is equally important. research is conducted on, with, or through oth-
Whether a researcher uses the deductive or ers. Thus, the communication researcher must
inductive process, the first research activity is seriously consider and evaluate the integrity of
to identify the research problem or topic. This the research proposed. Moreover, the research
process of identification is the focus of the must balance the scientific needs of the research-
next section of this chapter. Formulating the ers (as well as society’s need for knowledge)
problem into a research question or hypoth- with the physical, psychological, and emotional
esis formalizes the research as social science. needs of those who participate in the research.
Research conducted according to the deduc- These issues are addressed in Chapter 5. But,
tive model, which typically relies on quantita- for now, we need to discover how researchers
tive methods, is described in greater detail in identify the communication problem or topic of
Chapter 3. Research conducted according to interest.

key06767_ch02_018-034.indd 21 1/30/10 7:38:36 AM

 Rev.Confirming Pages

22 CHAPTER 2 / THE RESEARCH PROCESS: GETTING STARTED

IDENTIFYING THE RESEARCH is still important, however, because it can cer-

PROBLEM tainly lead you to the formal research question
or research hypothesis. Before you seek library
Remember that research is the process of asking resources, use the following steps to evaluate
questions and finding answers. Just as a detec- your preliminary question. Taking this step will
tive must examine the crime scene to get an idea help make your library search more effective.
of where to begin, you must identify a topic or an How does your topic lead to preliminary
issue as a focus for the project and the questions questions? Let’s say you are interested in the
that will follow. Identifying the research prob- impact of divorce on children’s abilities to com-
lem is always the first activity in the research municate their feelings. Closer examination
process. shows that there are two topics here: The first is
impact of divorce; the second is children’s abili-
Formulating Research Ideas ties to express their feelings. You could do library
research on each topic separately, but doing so
If you cannot think of a topic or problem, let might not lead you to the answer. By formulat-
your daily experiences guide you. What hap- ing your interest in this topic into a preliminary
pened today to you, or in front of you, that illu- question, “How does divorce affect children’s
minated a communication dilemma, problem, abilities to communicate feelings?” you are more
or question? What are your family, friends, and likely to seek resources that can answer your
colleagues talking about? How are their issues question or that will help you determine that the
related to communication? Answering ques- question has not been adequately answered.
tions like these can help you think of a topic or But take another look at that question. What
problem. Often we tend to think that the daily does the question assume? The question asks
problems of living we experience are unique to “how” divorce impacts children’s abilities. A
us. In reality, individual experiences may differ better first step would be to find out if divorce
in some ways, but generally they are connected affects children’s abilities to communicate. Thus,
to and mirror the experiences of others. Thus, “Does divorce affect children’s abilities to com-
consider whatever problems you are facing as municate feelings?” will be a better place to
a good source for research ideas. start, for it keeps you from falsely assuming that
Still unsure about a topic or problem to pur- divorce does impact children in this way.
sue? A good way to survey contemporary prob- As another example, you might recognize
lems and issues is to check what topics were on that the team leader of your shift at work has dif-
the front page of today’s newspaper or presented ficulty in organizing and conducting meetings.
on the radio or television. All the major news You wonder if there is anything you can do as
outlets—The New York Times, USA Today, ABC a team member to help. In this case, questions
News, CBS News, NBC News, CNN, and oth- could be, “Are leaders the only team members
ers—have web pages that are updated through- responsible for how meetings are conducted?”
out the day. or “In what ways can a team member maintain
the role of team member and help the leader con-
Turning Topics into Preliminary Questions duct more effective meetings?” or “What risks
do team members take when they help facili-
With the topic identified, you can begin to frame tate meetings?” Now look at these questions for
preliminary questions, which will help you the assumptions embedded within them. In the
search through library holdings and electronic first, you are asking about the basic assumption
databases. The preliminary question is not the of who is responsible for conducting team meet-
final research question you will see in both the ings. But notice in the second and third questions
deductive and inductive models. The research that the answer to the first question is assumed.
question is more formal, and the foundation for Rephrasing the topic as a preliminary
the research project. The preliminary question question is the first step in seeking answers.

key06767_ch02_018-034.indd 22 1/30/10 7:38:36 AM

 Rev.Confirming Pages

IDENTIFYING THE RESEARCH PROBLEM 23

Phrasing your question helps define your remember to formulate it into a question that
research area and narrow your search. Most focuses on communication. For example, a local
important, questions help you uncover the news item on the prevalence of sexual harass-
links between concepts and help you identify ment in your county government might end up
assumptions you have made. And, as they fre- as “In what ways does organizational culture
quently do, questions lead to more questions. If promote or inhibit the occurrence of sexual
you end up with several questions, try to order harassment?”
them into a list of which questions must be Take a look at the examples of topics and
addressed first, second, and so on. Or if ques- problems in the Try This! box “Developing Initial
tions in one area suggest questions in another Questions to Guide the Research Process.” When
area, try to draw a diagram of the relationship you have finished revising a few of the examples
of the questions to one another. Regardless listed, do the same to topics and problems that
of how you identified your topic or problem, interest you.

TRY THIS! Developing Initial Questions to Guide the Research Process

Read the example given in the table for the topic of instant messaging (IM). Notice
how the general topic or problem is stated as a preliminary question. Then the
question is analyzed for any underlying assumptions. For example, the sample
question assumes that all people have access to the Internet, use instant messag-
ing, and can explain why they use it. With these assumptions uncovered, the pre-
liminary question should be restated so that it is more specific. Use the topics and
problems listed in the following table to develop the preliminary questions to start
the research process.

Topic or State as Preliminary Examine Question(s) Restate Preliminary

Problem Question(s) for Assumptions Question(s)
Instant How do people All people have How do people
messaging use IMs? access to the explain their use
Internet. All people of IM?
who use the Internet
use IM. People will
be able to explain
why they use IM.
Parents talking
with their
children about
guns and
violence
Careers for
communication
graduates
Talking with
superiors at work

key06767_ch02_018-034.indd 23 1/30/10 7:38:36 AM

 Rev.Confirming Pages

24 CHAPTER 2 / THE RESEARCH PROCESS: GETTING STARTED

Evaluating Your Questions USING LIBRARY AND DATABASE

After you have developed your preliminary RESOURCES
question, it’s time for evaluation. Use these
questions to make a final assessment before you With preliminary questions developed, you are
spend time searching the research literature: now ready to conduct your search for resources.
It is tempting to conduct all your research online,
1. Is the question clearly stated? but working in the library has advantages as
2. Do others agree about the clarity of the well. The most important advantage: You can
question? ask a librarian for help.
If your library has the International Encyclope-
3. Have you asked only one question? Not dia of Communication or the Encyclopedia of Com-
two, three, or four? munication Theory in print or online, these are
4. What is the communication orienta- good first steps in looking for resources for your
tion of the question? In other words, preliminary question. The encyclopedic entries
what communication practice, policy, cover all aspects of communication. Each entry
rule, procedure, or consequence is being provides definitions, major themes of research,
investigated? Is your focus on symbols, and several key citations. Regardless of where
messages, or meanings? you gather resources, ask questions about the
5. Is the question phrased in such a way that it credibility, authority, relevance, and currency of
is not biased toward a particular answer or the source. How something is presented (in print
solution? or online) does not necessarily make a source
good or bad. If you select a source for your lit-
6. Is there some way to observe or measure the erature review or research project, you should
communication phenomenon of interest? feel confident that it is the best source for your
7. Can you research this question given the project. It is also important to gather several
limitations of time and resources? resources so you can compare definitions and
8. Who would be interested in the answer to research findings. The next section describes sev-
the question? eral ways to find resources for a literature review
or for a research project.
9. How could those who are interested use the
information?
Scholarly Journals
If you are satisfied that you are asking the
preliminary question in the most effective way Generally, your initial search should be for schol-
and that the question is appropriate for commu- arly articles published in academic journals.
nication research, you should identify the key- Articles in journals give you the opportunity
words or phrases to use in your search of library to see what research questions and hypotheses
resources. Taking the preliminary question have been studied by communication scholars.
above, “In what ways does organizational cul- You can read their arguments for the research
ture promote or inhibit the occurrence of sexual study they designed, and read how they col-
harassment?”, two keywords should be obvious: lected and analyzed their data. Most impor-
sexual harassment and organizational culture. What tantly, you can read about what they learned in
other keywords could also be useful? Sexual conducting the study.
harassment is a form of sexual discrimination or Journals are edited and published by scholarly
employment discrimination. The phrase organiza- professional associations, universities, or publish-
tional climate is often used to address work issues. ing houses dedicated to scholarly work. Scholars
To clearly tie your search for sexual harassment submit their manuscripts to a journal. The jour-
and organizational culture to the study of com- nal editor sends the manuscript out for review to
munication, you should include the keywords at least two reviewers who do not know the iden-
organizational communication or communication. tity of the author. This process allows reviewers

key06767_ch02_018-034.indd 24 1/30/10 7:38:36 AM

 Rev.Confirming Pages

USING LIBRARY AND DATABASE RESOURCES 25

to give their honest and critical feedback about Of course, there are other journals in the com-
the manuscript. After this peer review, the edi- munication discipline as well as journals that are
tor makes a decision about revision and publica- multidisciplinary (such as Business Communica-
tion. Often, journal articles are published only tion Quarterly, Critical Studies Critical Methodolo-
after an extensive review and revision process. gies, Journal of Business Communication, Journal of
In addition, most journals have a very high rejec- Contemporary Ethnography, Journal of Health Com-
tion rate, generally 80–90%. As a result, journal munication, Management Communication Quarterly,
articles are regarded as quality research written Qualitative Inquiry, Small Group Research). Finally,
by knowledgeable experts. Some of the journals journals in other disciplines (for example, man-
specific to the discipline of communication that agement, psychology, and sociology) do publish
publish social science research include research of interest to communication scholars.
In some cases, the research of communication
American Journal of scholars can be found there as well.
Communication Journal Communication Journals can be accessed in a number of ways.
Communication Journal of First, your library subscribes to article databases
Education Computer-Mediated that index communication research. Communica-
Communication tion & Mass Media Complete (often shortened to
Communication
CMMC) contains information on more than 600
Monographs Journal of Family
journals in communication, mass media, and
Communication Communication
related fields. Many times, the full text of jour-
Research Journal of Public nal articles are available through that database.
Communication Relations Research If an article you want is not in full text through
Quarterly Journal of Social and this or another database, look directly for the
Communication Reports Personal Relationships journal. Your library subscribes to print and
Journalism & online issues of hundreds, maybe thousands,
Communication
Communication of journals. Most library websites have a journal
Research
Monographs link where you can search for a specific journal
Communication title. That link will indicate if your library has
Research Reports Journalism & Mass
the journal in print or online, and what volumes
Communication
The Communication (or years) it has.
Quarterly
Review Accessing journals through your library’s
Mass Communication & website is recommended over performing a
Communication Studies
Society Google search for a journal. Why? Your library
Communication Theory has purchased journal subscriptions. Thus, there
Political Communication
Electronic Journal of is no cost to you for reading the online or print
Public Relations Review
Communication version. Check with your instructor or librar-
Qualitative ian to identify the journals that will satisfy your
Health Communication
Research Reports in needs during your literature search.
Howard Journal of Communication
Communications
Research on Language
Human Communication and Social Interaction Books
Research
Southern Your library also has a collection of books and
International Journal of Communication Journal edited books. Communication scholars often
Listening
Western Journal of write or edit books about their research exper-
Journal of Applied Communication tise. If you have identified an author who writes
Communication in the area you are conducting research, use his
Women’s Studies in
Research or her name in a search of the library catalog.
Communication
Journal of Broadcasting Or use one of your keywords as part of a title or
& Electronic Media subject area search in the library catalog.

key06767_ch02_018-034.indd 25 1/30/10 7:38:36 AM

 Rev.Confirming Pages

26 CHAPTER 2 / THE RESEARCH PROCESS: GETTING STARTED

You should also check to see if your library 1. What type of domain does the site come
carries Communication Yearbook. This annual from? Generally, .edu and .gov sites are
series is an edited collection of literature reviews considered more trustworthy than .org and
and topical critiques. For example, in Volume 32 .com sites.
(2008), Communication Yearbook published litera- 2. Who publishes or owns the site? How can
ture reviews on virtual teams and media literacy you tell? Look for this information between
among others. You will have to check the table http:// and the next /.
of contents of each Yearbook. If you find a chapter
3. Is this a personal website? One way to tell
that helps you, be sure to review the resources in
is to look for a tilde (~). Tildes often signify
the reference section.
a personal website. If a tilde is used with
Finally, check to see if your library has a
an organizational name, often what follows
handbook related to your area of research.
is the website of a specific person of that
Handbooks are focused on one context of com-
organization.
munication and provide extensive reviews and
critiques of literature, theories, and methods in 4. Can you tell who (a person or institution)
a particular area (examples are Handbook of Fam- created the site? Is there a name, e-mail
ily Communication, Handbook of Communication address, or a about us or contact us link?
and Aging Research, Handbook of Political Commu- Where do those links take you?
nication Research, Handbook of Interpersonal Com- 5. Are the author’s credentials or affiliations
munication, and Handbook of New Media). To find displayed on the website?
which handbooks your library carries, use the 6. What is the purpose of the website? To
keywords handbook and communication to search inform? Persuade? Sell? Is advertising
your university’s catalog database. If you find clearly labeled as such?
a chapter that helps you, be sure to review the
resources in the reference section. 7. How current is the website? Look for dates
to indicate when the site was created and
updated.
Online Resources
8. If a source is quoted or paraphrased on the
Because anyone can post a website and because web page, is information about that source
so many websites are posted, finding informa- provided so you can independently verify
tion on the Web is not a problem. But finding the accuracy of the information?
credible, authoritative information can be. You
will use a search engine (like Google) to search Anyone can produce a website. This means
material available on the Web. Remember that that you must carefully assess what is presented.
all search engines are not the same; each search A website can present research that appears to
engine has a different mechanism for finding be scholarly. But unless the site includes cita-
websites even when you use the same keywords tions and identifies the author and his or her
for searches. qualifications, it may be very difficult to gauge
There is nothing wrong with doing resource the authenticity or validity of the material.
searches on the Web. Online materials are not
necessarily unreliable. If you look online, out- Analyzing Your Resources
side of the university library website, you will
need to pay more attention to the credibility of After you have found several articles, books,
the resources you find. Like the journals and book chapters, or websites, you need to evalu-
books described previously, the first ques- ate these sources. Read through the abstracts of
tion to ask is: Is the online material written by each journal article or chapter you found, and
a credible and authoritative source? This list then read the discussion section. Read the fore-
of questions should help you make a decision word or introduction and first chapter of each
about the author’s credibility. book. Check the list of references at the end of

key06767_ch02_018-034.indd 26 1/30/10 7:38:36 AM

 Rev.Confirming Pages

USING LIBRARY AND DATABASE RESOURCES 27

the articles or chapters. Could any of the sources 8. What aspects of the topic have been
listed there be helpful to you? Are you able to ignored?
answer your question? Should your question be 9. Are there reasons to replicate, or repeat,
revised based upon what you have found? At studies that have been conclusive?
this point, you have reached another decision
10. What other topics have you found related to
point. If you can answer your question to your
your primary topic?
satisfaction, your search is over. If you cannot
satisfactorily answer your question, or if you If you have searched thoroughly and dili-
found conflicting answers in your search, you gently, you are likely to have uncovered the
can write your question in its final form. You are materials you need to answer your question
ready to develop your research project. or to develop your research project. Remem-
Answering the following questions can help ber, however, that it is nearly impossible to
determine if you have enough information or if find all the available literature. Finding every-
you need to continue on with a detailed search. thing is not a prerequisite for most student
Your library search has been adequate if you research projects. But you should have infor-
are satisfied with your answers to the following mation available from a variety of authors,
questions: from a variety of publication outlets, and
1. How much has been written on your topic? from sources published over time. Analyze
your resources for their breadth and depth of
2. How recent or relevant is the material? coverage.
3. Has some critical event occurred or societal As you review the literature you have found,
value changed that could challenge the take good notes and copy or print all the relevant
interpretation of the answers to the ques- pages. Check out the table of contents as well
tions asked? as subject and author indexes of books. Identify
4. Who has done the most work on your topic? books that are helpful by noting the authors’
names, complete book title, year of publication,
5. Where has research on the topic been
place of publication, publisher, and call number.
published?
Identify journal articles that are helpful by noting
6. What aspects of the topic received the most the authors’ names, complete article title, year
attention? of publication, complete journal title, volume
7. What questions about the topic have been number, and page numbers of the journal article.
answered? You will need all this information to develop

TRY THIS! Searching for Sources

1. Using the keywords media and ethics, perform a database search available
through your university library. Do the same keyword search on Google. How
would you describe and explain the differences?
2. Your question is “What nonverbal behaviors demonstrate confidence in public
speaking?” What keywords would you use in your basic search? Conduct this
search and report on your findings.
3. In your basic search for references on communication in stepfamilies, you have
found that communication scholar Paul Schrodt is the author of several studies.
How do you interpret this information? How would you use this information in
your detailed search?

key06767_ch02_018-034.indd 27 1/30/10 7:38:36 AM

 Rev.Confirming Pages

28 CHAPTER 2 / THE RESEARCH PROCESS: GETTING STARTED

a reference list if you cite the material in your Could several theories provide the basis for com-
research project. peting claims or solutions to your preliminary
question? As you find new sources of informa-
Adjusting the Question tion, it is likely that your preliminary question
will become more narrowly focused. Keep a list
As you work through the search strategies, don’t of all resources that you are using. You will use
hesitate to adjust your preliminary question. As these again as you develop your research project
you discover new information, you will develop and as you write up your research report.
a more sophisticated appreciation and under- It is time to stop adjusting the preliminary
standing of the problem. Incorporate the infor- research question when two conditions are satis-
mation you read into your preliminary question. fied. First, you should be comfortable that your
In particular, did your search uncover theories question is specific enough to be interesting. Sec-
that can help you make sense of your question? ond, you should be comfortable with the quality

AN ETHICAL Using the Ideas of Others

ISSUE
A detective relies on information and clues provided by others. So does a good
researcher. It would be impossible for one researcher to develop, document, and
validate everything one needed to know about a subject or topic. In research, then,
we must rely on the ideas and conclusions drawn by others. Anytime you use the
work of others, you must provide a citation indicating in the text of your paper what
idea you are using and whose idea it was. This is called an in-text citation. There
are two types. The first is the citation for a direct quotation. In this case, you indi-
cate with quotation marks the exact words you copied from the original work and
provide the page number in addition to the author’s last name and the year of the
publication. This way, anyone who reads your paper can locate the exact source.
Reel and Thompson (2004) conducted a three-phase study to investigate inter-
personal strategies of negotiating condom use. On page 102 of their article, they use
the following direct quote in-text citation.
In addition, Brown and Levinson (1987) tied the notion of face directly to “mention
of taboo topics, including those that are inappropriate” (p. 670).
The second type of in-text citation is for situations in which you have summarized or
paraphrased the ideas or conclusions of others. This in-text citation is documented
with the author’s last name and year of publication. An example of an indirect in-
text citation from the same article on page 100 is
Regarding condom negotiation, feelings of embarrassment are especially strong in
men (Cline & McKenzie, 1994); in particular, men are especially likely to feel that
suggesting condom use will diminish the chances of sexual intercourse (Bryan,
Aiden, & West, 1999).
A complete list of your references is provided at the end of your paper. This enables
the reader to locate the source of any in-text citation you used. Most social science
researchers use the citation and reference style of the American Psychological Asso-
ciation (6th ed.), which is the style used in this book. Check with your instructor to
see which style you should use.

key06767_ch02_018-034.indd 28 1/30/10 7:38:36 AM

 Rev.Confirming Pages

USING AND ORGANIZING YOUR RESOURCES 29

DESIGN Evaluating the Literature You Found

CHECK
As you can see, the research process starts with your identification of a research
topic and your search of the research literature. You will use the articles and chap-
ters that you find in the literature review for your study. Communication research
is indexed on several databases, so if you are not finding what you need, be sure to
ask your reference librarian for help. As you search the literature, be sure that you
are collecting studies that are published in communication journals or are authored
by communication scholars. Scholars in many disciplines study communication,
but the most complete and thoughtful focus on communication is published by
communication scholars.

and quantity of resources you can use to help investigation is found in the discussion section.
you answer your question. At this point, it is When a research question or hypothesis is pre-
time to move on to analyzing the resources you sented in the literature review, it is still tentative.
have collected.
Tracking Others’ References
USING AND ORGANIZING YOUR As you read the articles, books, and chapters,
RESOURCES you will find in-text citations. This documenta-
tion device provides information within paren-
Getting started on a stack of resources is not easy. theses for the research work cited by the author.
The first step is to read the title and abstract or Each in-text citation includes the authors’ last
summary of each resource. Next, read the prob- names, year of publication, and the page num-
lem statement, which is usually part of the litera- ber if material is quoted word for word. To
ture review or precedes the literature review in track down this citation, turn to the reference
a journal article or book chapter. It identifies the list at the end of the article, book, or chapter.
research objectives. Although the exact research This is labeled with the heading “References,”
questions or hypotheses may not be presented “Bibliography,” or “Works Cited.” For each
here, the problem statement generally sug- citation you will find a complete bibliographic
gests them. The problem statement answers the entry—all the information necessary for you to
question “Why did the researchers conduct this find the article, book, or book chapter. Look at
study?” Generally, reading this section will help the “References” section at the end of this book
you decide if the article or research report will be for an example.
helpful to you. Why would you want to track down the
In the literature review, the authors present articles, books, and book chapters that other
the literature that supports their formal research authors have used? There are several reasons.
questions and hypotheses. Read the research First, these published works are part of an
questions and hypotheses carefully because the ongoing scholarly conversation. Something
results or conclusions from the study are tied briefly mentioned in one article might lead you
directly to them. For now, skim the methods and to another article that could provide valuable
results section, and then move on to the discus- background information for you. Second, you
sion section. What did the scholars find? What may have missed this source in your library
were the answers to the research questions? Did search. Tracking down the references used by
they confirm or not confirm the hypotheses they others gives you the opportunity to fill in the
proposed? Remember that the conclusion to the gaps of your literature search. Third, authors

key06767_ch02_018-034.indd 29 1/30/10 7:38:36 AM

 Rev.Confirming Pages

30 CHAPTER 2 / THE RESEARCH PROCESS: GETTING STARTED

draw conclusions about the work of others and question?” If you have, you will need to think
then base their arguments on those conclu- creatively about how to replicate, extend, or chal-
sions. If you are not familiar with the literature, lenge those conclusions as the basis for a research
you have to take the authors’ conclusions for project. Generally, there is little value in repeat-
granted. Rather than relying on their evaluative ing the work of others if you agree with their
biases, you could track down the reference, read conclusions and find no major faults in how they
it, and draw your own conclusions. conducted the research. However, there is value
in replication when the original study is dated,
Organizing What You Have Found when societal values and practices surrounding
the issue have changed or are changing, or when
Now, how will you organize the resources you you find a flaw in the study that makes you
have found so you can write a research report question the results. For example, societal values
or a research proposal? First, look at the body of about many adult relationships (such as mar-
literature you have collected. Are there one or riage, living together, divorce, single parenting,
two authors whose names appear several times? adoption) have changed significantly. Research
If so, start your reading there. Researchers tend in these areas may be necessary to see if the con-
to work on lines, or streams, of research. This clusions drawn in the past about communication
means that scholars become known for conduct- in these relationships are relevant now.
ing research on certain topics. If the names of one As you read and sort the literature you found,
or two scholars don’t stand out, organize your you might find that scholars disagree. Or you
literature by publication date. To get a historical might find that research conclusions have been
overview, read from the older literature through drawn about most, but not all, of the issues sur-
to the newer literature. Another way to start is to rounding your topic. If scholars disagree, you
begin with an article or chapter that reviews or could develop a research study to examine the
summarizes a particular line of research. If you disagreement. If some, but not all, issues are
find conflicting ideas, theories, or findings, orga- answered, you could develop and conduct the
nize resources into similar categories. study that fills in this gap. Remember that one of
the characteristics of science is that it is heuristic.
Summarizing What You Have Found This means that conclusions, or answers to ques-
tions, help identify new questions to be answered.
One method is to arrange the material using You now have the basis of the literature review
major and minor points as primary and second- section of your research report. The next step is to
ary headings, much like a traditional outline. develop the formal research question or hypoth-
A second method is to arrange the findings in esis and design your research study. Chapter 3
chronological order, usually working from the will give you more information for developing
oldest to the most recent. This is particularly a quantitative study. Chapter 4 will do the same
helpful if you want to demonstrate how a ques- for developing a qualitative study. When you are
tion was answered or how a topic developed ready, you can find information about writing
over time. A third method is to ask a series of literature reviews in Chapters 17 and 18.
questions and respond with what you found for
each question. In this case, working from the
broadest question to the narrowest question is CONSIDERING THEORY
recommended. A final method for organizing IN RESEARCH
your material is to work from general to specific
(a deductive approach), or build from the spe- Research revolves around theory. Thus, the jour-
cific to the general (an inductive approach). nal articles and book chapters you find in the
Whatever approach you take to summarizing library use research as a basis for developing or
what you found in your literature search, the pri- challenging theory. Theory is a set of interrelated
mary question should be—“Have I answered my propositions that present a systematic view of

key06767_ch02_018-034.indd 30 1/30/10 7:38:36 AM

 Rev.Confirming Pages

CONSIDERING THEORY IN RESEARCH 31

phenomena. The purpose of those propositions Developing Theory

is to describe, predict, or explain the phenom-
ena. For communication research, theory cre- We engage in informal theorizing when we try
ates propositions about symbols, messages, and to make sense of the past, operate effectively in
meanings. the present, or anticipate events in the future
Research is necessary to validate theory. Gen- (Lustig, 1986). Although theorizing is a common
erally, quantitative research starts with a theory. and fundamental human activity practiced every
Then researchers conduct a research study to day, formal theorizing as a scientific process is
demonstrate if a theory holds true for a set of data. quite different. Formal theory building (Lustig)
If it does not, the theory is altered or discarded. involves six basic steps:
In this theory–research link, theory is the map In step 1, the researcher describes an event
by which the researchers conduct their studies. or observation that needs understanding. The
As described earlier, this type of research relies event must be observable, interesting, impor-
on deductive thinking or reasoning in that theory tant, and remarkable to come to the attention
presumes what will result and the research veri- of the researcher or someone else who desires
fies those claims. Theory directs the researcher an understanding of it. This first step begins to
in developing hypotheses and questions and in identify the “what.”
selecting the method for testing them. In step 2, the researcher creates an explana-
Research is also necessary to develop theory. tion for the event. Although anyone can create
Generally, qualitative researchers start with a an explanation, it is the scientist’s job to formal-
research question and use their findings to both ize and test explanations. In this step the answer
answer the question and contribute to theory to “Why?” begins to be formulated.
development. In this theory–research link, the In step 3, the researcher moves from the spe-
theory, or map, is drawn from the experiences cific event or observation to a more generalized
uncovered by the research. This type of research form. In other words, if the event of interest is
relies on inductive thinking or reasoning in that family decision making around the dinner table,
theory, or generalization, is derived from the the researcher could move to the more general-
cases explored. ized communication event of decision making
As you can see, research and theory are nec- or the more generalized communication event
essary complements to one another. Theorizing of family interaction. The researcher must decide
is important to research in two additional ways which type of communication event is more
(Brooks, 1970). First, researchers cannot observe interesting and intriguing to investigate. By mov-
the entire universe. Rather, researchers must ing to a more abstract level, the researcher can
select a subset of phenomena to be observed. now look for similar events to see if the answer
Theory directs researchers’ attention to particu- to “Why?” developed in step 2 is also suitable
lar communication patterns, functions, themes, for explaining these other, different but simi-
processes, and so on. For example, it would be lar, events. Instead of focusing on one specific
impossible for a communication researcher to interaction event, the researcher must develop
study every aspect of how communication is answers suitable for a class of similar events. This
used in political campaigns. Thus, theory helps us characteristic of theory moves it from an informal
define and isolate a communication phenomenon to a formal level. Although you are comfortable
for study. Second, theory helps “integrate data with the way informal theorizing describes and
which otherwise would remain mere collections explains events that happen in your daily life,
of facts” (Brooks, p. 4). This issue returns us to you would not be comfortable applying others’
the definition of theory given earlier. In a theory, informal theories to the events that you experi-
research findings are integrated into a system of ence. Thus, the researcher’s job is to discover the
description, prediction, and perhaps explana- commonalities among events that allow them to
tion that help us answer questions of “What?” be classified and then to develop and test theories
“Why?” “How?” and sometimes “What if?” that describe and explain all events belonging to

key06767_ch02_018-034.indd 31 1/30/10 7:38:37 AM

 Rev.Confirming Pages

32 CHAPTER 2 / THE RESEARCH PROCESS: GETTING STARTED

TABLE 2.1 Theory Development—Steps 1 Through 4

Task Example
Step 1 Describe event or Family members (2 adults, 2 children) eat dinner and discuss
observation their daily activities. Father introduces family activity for
weekend, which generates considerable discussion from chil-
dren. Although the discussion initially has both positive and
negative points introduced, eventually the children agree that
they do not want to pursue the weekend activity suggested.
Step 2 Create explanation Explanation 1: Children are likely to reject ideas presented by
for event parents during dinnertime discussions.
Explanation 2: Parents introduce ideas for family dinnertime
discussion to obtain family members’ preferences.
Step 3 Move from specific General form 1: Children’s rejection or acceptance of parental
to more generalized input. General form 2: Parents desire input from other family
form members.
Step 4 Derive predictions Focus 1: Children are likely to reject ideas presented by
from explanations parents.
Focus 2: Parents will seek input about family matters from
other family members.

a class. Thus, a theory of decision making should proposition, the discrepancy must be explained
apply to many people’s experiences of decision by critically examining the methodological pro-
making, not just one’s own. cess or by reworking the theoretical framework.
In step 4, the researcher begins to derive If the theoretical framework is revised or if two
predictions from the explanation developed in alternative and competing explanations are pres-
step 3. To do this, the researcher asks, “What ent, the theory-building process starts again. If
else would be true, or observable, if the expla- methodological problems are identified, the
nation was correct?” Continuing with our family researcher repeats steps 5 and 6 using different
decision-making example, the researcher could and improved methodological procedures.
make several propositions that are testable. Even after these six steps, the theory-building
Examine Table 2.1 to see the progression from process is not complete or final. Theory is devel-
step 1 through step 4. oped over time, and this theory-building process
Now, in step 5, the researcher must select a is repeated many times as different scholars test
focus and test the proposed theory. Most com- theoretical propositions in their research. Both
munication observations are complex enough quantitative and qualitative research contribute
to support multiple attempts at theory building. to theory development.
The researcher must develop a plan for and col- Theories are developed and tested incremen-
lect data that can test the predictions. tally. After a basic theoretical notion is presented
Step 6 of the theory-building process uses the as a proposition in the scholarly literature, schol-
obtained data to confirm, revise, expand, gener- ars develop studies to test the propositions.
alize, or abandon the proposition tested (Lustig, This is possible because the results of scholarly
1986). Notice that collecting the data in step 5 is research are presented in a public forum. Theory
distinct from interpreting the data in step 6. If the is confirmed only after many studies, usually
results are consistent with the proposition, the conducted by different scholars with different
theoretical framework is confirmed for the time methodologies, achieve similar results. Even at
being. If the results are not consistent with the that point, theories are still considered tentative.

key06767_ch02_018-034.indd 32 1/30/10 7:38:37 AM

 Rev.Confirming Pages

CONSIDERING THEORY IN RESEARCH 33

TRY THIS! Finding Theory in Journal Articles

Find two or three journal articles for a communication topic that interests you. Care-
fully read the literature review of each article. Does the author identify by name the
theory or theories that are providing the foundation for the research study? Does
the author point to a description, cause, prediction, or explanation as the reason for
conducting the research? If so, this is likely the theoretical basis of the study. Next,
read the discussion and implication sections of the articles. In this part of the jour-
nal article, authors discuss the implications of the study as a challenge to the theory
or as further development or expansion of the theory.

A theory that was at one time believed to be valid research could produce results without an orga-
can be questioned in the future. For example, new nizing framework. Second, theory links one
technologies can create new opportunities and study to another, helping to provide a continual
circumstances for communication. Thus, theo- conversation about our understanding of com-
ries of how and why interpersonal relationships munication phenomena. Third, theory provides
develop over time may need to be reexamined a framework for understanding how concepts
in light of the extent to which these technologies and issues are important or significant in our
are used in developing relationships. interactions. For example, theorizing about com-
munication apprehension and then conducting
Utility of Theory–Research Link studies to validate those expectations helped
researchers uncover the role apprehension plays
To the extent that a community of scholars in nearly every communication event in which
accepts research findings and can agree on we participate. Fourth, theory helps us inter-
the theoretical propositions, theory has been pret the larger meaning of research findings. For
achieved. But all theory should be judged by example, reading about how observers react to
some aspect of utility (Lustig, 1986). The knowl- an apprehensive individual may cause you to
edge gained from the process of theory-building monitor and manage your own apprehensive-
should be used “to suggest new questions that ness when speaking in public.
are worth answering, develop more accurate Scientific and systematic inquiry is a process
theories about human communication, com- of developing and testing theory. Direct relation-
municate more effectively, teach others to com- ships exist among questions asked, data observed,
municate more effectively, create better human and theory development (Miller & Nicholson,
relationships, and improve the cultures and the 1976). Examine the deductive and inductive
environments within which we all live” (Lustig, research models (see Figures 2.1 and 2.2). See
p. 457). When the utility criterion is added as a how theory drives quantitative research? Alterna-
test of the theory-building process, you can see tively, in qualitative methodology, observations
not only that the theory–research relationship tend to drive theory development. However, the
is reciprocal, but also that it is grounded in the selection of any particular quantitative or quali-
practical issues of human communication (Key- tative methodology does not guarantee that a
ton, Bisel, & Ozley, 2009). study will result in theoretical development.
In fact, theory is used four ways in the Rather, a study must be designed to illuminate
research process (Hoover & Donovan, 1995). and examine underlying principles and proposi-
First, theory provides patterns for interpreting tions (Shapiro, 2002). Only then can its findings
data. Without working from or toward theory, contribute to theoretical development.

key06767_ch02_018-034.indd 33 1/30/10 7:38:37 AM

 Rev.Confirming Pages

34 CHAPTER 2 / THE RESEARCH PROCESS: GETTING STARTED

Also recognize that the process of inquiry is 9. Evaluate your preliminary questions for
not always linear. Nor can theory be developed their underlying assumptions, complete-
or challenged in one study. Recall that science ness, and clarity prior to conducting the
can be characterized by its replicable and self- library search.
correcting nature. Multiple studies are needed 10. Find scholarly articles published in aca-
to replicate findings, just as multiple studies are demic journals or scholarly books.
needed to challenge and alter existing theory.
11. When you find an article or book that may
be helpful, take notes and document all the
citation information.
SUMMARY
12. Your preliminary question may require
adjustment as you discover new
1. Researchers seek answers to questions.
information.
2. Library research can reveal if the answer to
13. The abstract states the primary ideas pre-
your question is available, but not known
sented in an article or chapter.
to you.
14. The literature review usually concludes
3. If the answer is neither known nor avail-
with the researchers’ formal research ques-
able, research must be conducted to
tions and hypotheses.
uncover the answer.
15. The discussion section includes the answers
4. Research can be a deductive or an inductive
to the questions the authors raised.
process.
16. Using what you found in the library search,
5. The steps of the deductive process are
organize your material by major and minor
identifying the research problem, review-
points, in chronological order, by answering
ing existing theory, formulating a research
a series of questions, or from the general to
question or hypothesis, designing the meth-
the specific or from specific to general.
ods and procedures, gathering data, and
analyzing and interpreting data. 17. Theory is developed and tested through
research.
6. The steps of the inductive research process
are identifying the research problem, con-
sidering existing research findings, devel-
oping the research question, designing the
KEY TERMS
methods and procedures, gathering data,
analyzing and interpreting data, and devel- deductive inductive
oping a theoretical explanation.
7. Both the deductive and inductive research
processes are circular and cyclical as the
final step, asking new questions, starts the See the website www.mhhe.com/keyton3 that
research process again. accompanies this chapter. The site contains
8. The first step in both the deductive and a chapter summary, outline, and checklist. A
inductive research processes—identifying PowerPoint presentation created by Dr. Keyton
the research problem—consists of identify- is also available for download, as well as a short
ing the topic or issue, turning the topic into multiple-choice quiz to check your learning.
a preliminary question or set of questions, You will also find a list of Internet resources that
conducting a library search, and adjusting provide further information about the issues in
the question, if necessary. this chapter.

key06767_ch02_018-034.indd 34 1/30/10 7:38:37 AM

 Rev.Confirming Pages

CHAPTER THREE

Introduction to Quantitative
Research

Chapter Checklist

After reading this chapter, you should be able to: 8. Distinguish among concepts, conceptual
schemes, constructs, variables, and
1. Describe quantitative research and its
operationalizations.
assumptions.
9. Identify independent and dependent
2. Identify examples of quantitative research.
variables.
3. Explain analytic deduction.
10. Explain the relationship between
4. Explain the five-component model for independent and dependent variables.
quantitative research.
11. Explain the advantages and disadvantages
5. Explain the role of hypotheses in of quantitative research.
quantitative research.
12. Describe issues of reliability and validity
6. Assess the effectiveness of hypotheses in that must be addressed in quantitative
quantitative research. research.
7. Explain why research questions are used in
quantitative research.

key06767_ch03_035-056.indd 35 1/25/10 6:51:48 PM

 Rev.Confirming Pages

36 CHAPTER 3 / INTRODUCTION TO QUANTITATIVE RESEARCH

When you think of research, you are likely to about the use of profanity in films, whereas
think of a laboratory experiment, which is what others question whether the regulations are still
has traditionally been associated with scientific needed. Cressman, Callister, Robinson, and Near
research. In communication research, lab experi- (2009) designed a quantitative research study to
ments are only one of a variety of quantitative identify what types of profanity were used and
methods from which a researcher can choose. their frequency. The research team selected 30 of
This chapter provides a basic introduction to the top grossing teen films from the 1980s, 1990s,
quantitative communication research. You will and 2000s that had received G, PG, and PG-13
discover how quantitative methods rely on the ratings. G-rated films should not contain offen-
identification of variables and the development sive language; PG-rated films suggest parent
of testable hypotheses and questions. Moreover, guidance because these films can contain some
you will discover that the way in which the profanity; and PG-13-rated films caution parents
research question or hypothesis is written actu- because some of the language in the film may be
ally helps the researcher in selecting the research inappropriate for children under 13.
method. But do not conclude that quantitative How were quantitative methods used in this
research can answer all of our questions about study? To answer their research questions and
human communication behavior. Some ques- hypotheses, the research team counted frequen-
tions are better answered by qualitative research cies of occurrence for language in the films that
methods, which are introduced in Chapter 4. could be described in one of five categories: the
seven dirty words that the FCC deems unspeak-
able for broadcast, sexual words, excretory
WHAT IS QUANTITATIVE words, strongly offensive, or mildly offensive.
RESEARCH? Across the 90 films coded, 2,311 instances of
profanity were identified. Teen characters spoke
As the label implies, the unit of analysis in quan- about two-thirds (69.1%) of these words; teen
titative research is quantity (Anderson, 1996). males used more profanity than teen females or
Researchers use measurement and observation adults of either sex. While mild profanity was
to represent communication phenomena as identified more frequently (57.1%), the seven
amounts, frequencies, degrees, values, or inten- dirty words were the second highest category
sity. After phenomena are quantified, research- (22.1%). By first identifying and then interpret-
ers compare or relate them using descriptive or ing the frequency of occurrence of language in
inferential statistics. By using traditional quan- the different categories of profanity, the research
titative approaches and statistical techniques, team was able to demonstrate that there was a
researchers bring greater precision and, as a slight decrease in the use of profanity across the
result, some would argue, greater objectivity to three decades. However, the use of profanity
the study of communication phenomena. A few was still prevalent in the most recent set of films.
examples will demonstrate the variety of quanti- Thus, the findings question the influence of the
tative research methods available to communica- film rating system on films that teens will see.
tion researchers. Other quantitative studies are conducted with
questionnaires or surveys. Advances in technol-
Examples of Quantitative Research ogy have changed the ways businesses oper-
ate and how job applicants interact with them.
In part, film ratings (i.e., G, PG, PG-13, R, NC-17) To understand the preferences of employers
in the United States are based on characters’ use in receiving resumes and cover letters, Schull-
of language, especially profanity. Many parents ery, Ickes, and Schullery (2009) sent surveys by
rely on these ratings to make appropriate media e-mail to all the organizations that recruit at their
choices for their children. Although the rating university. Responses were received from 140
system went into place in 1968, parents, media nationally based U.S. companies and 92 multina-
researchers, and policy makers still have concerns tional companies.

key06767_ch03_035-056.indd 36 1/25/10 6:51:48 PM

 Rev.Confirming Pages

WHAT IS QUANTITATIVE RESEARCH? 37

AN ETHICAL Giving Your Children Permission to Participate in a Research Study

ISSUE
To collect data from participants, researchers must ask for and gain the consent
of participants. In addition to gaining participants’ consent, researchers need the
consent of a parent or guardian before individuals under the age of 18 can par-
ticipate in an academically sponsored research project. If you were the parents of
a 13-year-old, would you give your daughter or son permission to participate in a
research project on dating? Drinking? Drug use? Why would some parents object
to their teenager’s participation? Is there any way the researcher could overcome
these objections? What are the potential benefits of having teenagers participate in
research projects like these?

Purposely avoiding the terms online, web- us by viewing our Facebook page, for example?
based, or electronic, the survey asked organiza- Would their impressions change if others posted
tional representatives to provide answers to comments about our physical attractiveness that
questions about the type of resume they prefer were different from how we describe ourselves?
to receive, how they prefer to receive resumes, Walther, Van Der Heide, Harnel, and Schulman
and if they prefer a cover letter with the resume. (2009) investigated this question with an experi-
Each question could be answered with a set of ment. Using mock-ups of a Facebook profile
response options. Frequency counts for each page, the research team was able to manipu-
response were reported. The research team late the information the participants saw. Some
found that regardless of the company’s size, type, participants saw self-statements or descriptions
geographic range, or the job function of the job posted by the owner of the profile page; other
applicant, employers preferred that resumes be participants saw statements posted by the pro-
formatted in the traditional chronological order. file owner’s friends. Some of the friends’ state-
Companies did differ in how they preferred to ments described the profile owner as attractive
receive resumes. Generally, companies preferred (e.g., “If only I was as hot as you); other state-
to receive resumes as e-mail attachments (46%) ments described the person in unattractive ways
or through their websites (38%). Only 7% pre- (e.g., “Have you lost weight? Good job keep
ferred receiving resumes by mail. These prefer- workin’ at it”). Research participants provided
ences were not related to geographic area or type evaluations of the profile owner’s attractiveness
of company. But smaller organizations did pre- and their honesty. In terms of attractiveness,
fer to receive resumes by e-mail, and resumes for comments made by the profile owner’s friends
positions in human resources were preferred to were more influential. That is, friends’ descrip-
be received through company websites. Interest- tions of attractiveness were more trusted than
ingly, a little more than half (56%) of the compa- the physical descriptions provided by profile
nies preferred to receive a cover letter with the owners.
applicant’s resume. Using surveys or question- In another experimental design, Ivory and
naires is a very popular quantitative method. Kalyanaraman (2009) asked participants to think
Chapter 9 explores this topic in detail. of and then write down the most violent video
Communication researchers also use experi- game with which they were familiar; another set
mentation to capture quantitative informa- of participants were not asked to identify a video
tion about communication issues. Most of us game. Next, participants were asked to rate the
use social networking sites, and, in our pro- degree to which they believed the specific video
files, most of us describe ourselves in favorable game, or video games in general, would cause
ways. But what impressions do others have of others to be aggressive or violent. The research

key06767_ch03_035-056.indd 37 1/25/10 6:51:48 PM

 Rev.Confirming Pages

38 CHAPTER 3 / INTRODUCTION TO QUANTITATIVE RESEARCH

team wanted to test the hypothesis that “partici- because they are a direct result of testing the
pants’ perceptions of violent video games’ effects theory. Or the methods or procedures followed
on aggression will be greater when considering a could be faulty. In this case, the researcher devel-
specific game compared to violent video games ops a new research design for testing the theo-
in general” (p. 4). The research findings did not retical propositions and collects new data.
support the hypothesis. Rather, just the opposite
was found. Participants who named a specific A Model for Conceptualizing
video game believed that the game would cause
Quantitative Research
others to be less aggressive or violent than those
participants who were not asked to identify a The deductive research model presented in
video game. Chapter 2 is a general overview of the research
The point of this experiment was to under- process: It gives the basic steps and can be
stand what factors influence people’s percep- applied to most quantitative research projects.
tions about the effects of violent video games. But with both quantitative research and qualita-
It was an interesting experiment because the tive research (see Chapter 4), more specialized
research team did not have participants view or models are needed to guide us through selecting
play any video games. Rather, the manipulation and developing the research plan for a specific
was invoked by simply asking some participants study. The model for conceptualizing quantita-
to think of and write down the title of a violent tive research shown in Figure 3.1 is the model
video game. Thus, the results demonstrate that that will guide us now.
violent video games in general caused partici- Starting at the top left of the model, the first
pants to believe that video games were more component is the research purpose. Because the
likely to cause aggression and violence. Alter- researcher is familiar with the research literature
natively, when participants wrote down the title and has developed some questions about com-
of a violent video game, their perceptions about munication issues, she or he begins with an
potential aggression and violence were lower. overall purpose or objective. For example, I am
These are just four examples of ways in which interested in how a dysfunctional group member
communication researchers used quantitative can take over a group’s interaction to the point
methods. One element underlying these exam- that the ineffective member replaces the task as
ples that may not be apparent to you, but that the focus of the group. So my research purpose is
needs discussion, is the pattern of reasoning used to study that communication phenomenon.
by these researchers. We examine that next.

Deductive Reasoning
Purpose Literature
Quantitative research relies primarily on deduc-
tive reasoning (Hawes, 1975). This means that
the researchers select a theory, or theories, as the
basis of the propositions that are tested in the
study. In this case, the logic flows from the gen- Research questions
eralized (the theory) to the specific (the research and hypotheses
conclusion). In general, the researcher hopes that
the research process supports, or verifies, what
the theory proposes to be true.
If the data and results do not support hypoth- Validity/
eses derived from the theory, then the researcher Methods
reliability
looks for an alternative explanation. Perhaps
the theory, as it was developed, is deficient or FIGURE 3.1 A Conceptual Model for Quantitative Research
incomplete. Therefore, the results must be faulty

key06767_ch03_035-056.indd 38 1/25/10 6:51:48 PM

 Rev.Confirming Pages

WHAT IS QUANTITATIVE RESEARCH? 39

Notice that the purpose component alone communication phenomena related? Either of
does not drive the model. The traditions of social these two forms is acceptable, but they require
science require that researchers use literature as different types of statistical approaches. If I want
a basis for their research, so my desire to study to study the difference between dysfunctional
dysfunctional group members is framed within members and functional, or effective, members,
the context of the research literature and the then I choose a quantitative method that will
contexts of real groups that I have observed help me examine and illuminate those differ-
experiencing this phenomenon. I take informa- ences. If I want to study how the dysfunctional
tion (and motivation) from my original purpose, member influences the group’s decision mak-
the research literature, and my experiences with ing and conflict management, I would choose a
groups. Balancing what I know as an academic method that illuminates the relationships among
and what I know as a group facilitator leads me the degrees of dysfunctional, decision-making,
to the research questions or research hypotheses and conflict management behaviors.
that will guide my study and selection of quanti- Looking at some specific examples will help
tative methods. you distinguish between instances in which
Research questions and research hypotheses differences are the focus of the research and
are central to the quantitative research process. instances in which relationships between vari-
No quantitative study can be done without one ables are the focus. Think about your interaction
or a combination of these. This third component with your significant other and your interaction
dominates the quantitative research methods with your supervisor. First, think about how the
process. Later in this chapter we will explore two relationships are different. Look at Table 3.1
these in greater detail. But, for now, recognize to see what I mean.
that without one or several research questions or In comparing the interaction with a signifi-
research hypotheses, there is nothing to direct or cant other to the interaction with a supervisor,
define the research process. it is easy to identify several variables on which
The fourth component of the quantitative there will be differences (for example, status,
research model is the selection of the research motivation, and level of intimacy). Notice how
methods for the project. Generally, quantita- a difference can be extreme, as in the status and
tive researchers ask questions about differences motivation examples. Also notice how the dif-
and relationships. In other words, how are com- ferences may be more moderate, as in the level
munication phenomena different? Or how are of intimacy of the conversation. In each of these

TABLE 3.1 Looking for Differences Between Two Types of Interaction

Variables of Potential Differences

Between the Two Relational Types Interaction with Significant Other Interaction with Supervisor

Relative status of other person I consider this person to be my Due to the organizational
in the relationship equal in terms of power and hierarchy, this person has
status. more power and status in the
organization than I do.

Motivation for relationship We are friends. I voluntarily I was assigned to my

developed this relationship. supervisor’s work unit. This
relationship is not voluntary.

Level of intimacy in We share personal information I do share some personal information

conversations with one another. with my supervisor, especially
when we eat lunch together.

key06767_ch03_035-056.indd 39 1/25/10 6:51:49 PM

 Rev.Confirming Pages

40 CHAPTER 3 / INTRODUCTION TO QUANTITATIVE RESEARCH

TABLE 3.2 Looking for Relationships Between Two Types of Interaction

Potential Relationships Between

the Two Relational Types Interaction with Significant Other Interaction with Supervisor

Number of conflicts I frequently have conflicts with my I frequently have conflicts with
significant other. my supervisor.

Degree of satisfaction with I am very satisfied with my interaction I am not satisfied with my
interaction with my significant other. interaction with my supervisor.

Frequency of interaction My significant other and I talk very I talk to my supervisor only
frequently. when necessary.

cases, the researcher would quantitatively mea- Returning to my example of dysfunctional

sure status, motivation, and level of intimacy for group members, if I focus on differences, I might
each participant twice—once for each relation- look for ways in which functional and dysfunc-
ship type. Then the researcher would compare tional group members deal with conflict, facili-
the scores for interaction with the significant tate the group’s decision making, or challenge
other to the scores for interaction with the super- the group’s leadership. If I focus on ways in
visor to see if a difference existed. which these two types of group members are
Alternatively, the research project could be related, I might want to focus on the degree to
designed to explore how the two types of inter- which each type of group member possesses
actions are related to one another. Examine communication competence, or how similar the
Table 3.2 to find these examples. In looking for types of group members are in their communi-
how interaction in the two relational types is cation style. To highlight differences, one set of
related or similar, the researcher would again statistical techniques is used. To highlight rela-
obtain a quantitative measurement of number tionships, another set is chosen. These statistical
of conflicts, degree of satisfaction, and frequency techniques are covered in Chapters 11 and 12.
of interaction for each relational type. Then the Notice that the final component of the quan-
researcher would examine the pairs of scores to titative research model is connected only to the
see how they relate to one another. Notice how methods component and that the connection is
the scores may be very similar, or highly related, reciprocal (see Figure 3.1). This last component is
as with the number of conflicts. Or the scores can an examination of the validity and reliability of
be related, but in opposite directions, as with the data collected through the method selected.
satisfaction with interaction and frequency of Usually the researcher will make methodologi-
interaction. cal choices and then assess those choices for their
Thus, we can look for the differences between impact on validity—how truthful the data will
these two relational types, or we can look for be—and reliability—how consistent the data will
ways in which interaction in the two relational be. These issues must be addressed before the
types is related. The point here is that almost any research is undertaken and any data collected, for
communication phenomenon can be examined there is little opportunity to adjust the research
for its differences to a similar communication method after the project is started. One of the
phenomenon, or a communication phenomenon assumptions of quantitative methods is that all
can be examined for how it relates to a similar participants are treated similarly; procedures or
communication phenomenon. Can you think of process should not change as you discover errors
other elements that may differ or be related with or lapses in your planning.
respect to relationships with significant others After the researcher has addressed each
and supervisors? of these five components, he or she moves

key06767_ch03_035-056.indd 40 1/25/10 6:51:49 PM

 Rev.Confirming Pages

CREATING THE FOUNDATION FOR QUANTITATIVE RESEARCH 41

through the quantitative research process in a consequence is demonstrated in prime-time

linear fashion. Once the methods are designed, television dramas (for example, characters admit
the researcher selects participants, collects and guilt and take responsibility, deny responsibil-
analyzes data, and then writes the research ity, assign responsibility to others, and so on).
report. As demonstrated in Figure 2.1, find- Individually, each concept describes a unique
ings from the current study are used to extend process. But as a group, the concepts still retain
or challenge the current state of theory. New common characteristics. Together, they form a
questions are formulated and the research pro- conceptual scheme that specifies and clarifies the
cess begins again. For now, though, we need to relationships among them (Kibler, 1970).
return to and examine in detail each step in con- The theoretical definition of a concept is a
ceptualizing the research project because these construct. Concepts can become constructs only
steps provide the foundation for quantitative when they are linked to other concepts. The link-
research projects. ing between and among concepts is part of the
theoretical definition. To be used in a research
study, however, a construct must also be assigned
CREATING THE FOUNDATION an observable property, or a way for a researcher
FOR QUANTITATIVE RESEARCH to observe or measure it. Obviously, a construct
could be observed or measured in many ways.
After identifying the research problem and turn- Thus, researchers use the term variable to iden-
ing that topic into a preliminary question or tify the theoretical construct as it is presented in
questions, as described in Chapter 2, researchers research questions and hypotheses, and the term
must further specify the concepts identified in operationalization to denote how the variable
their question. These conceptual definitions are is observed and measured. See Figure 3.2 for a
based on theoretical information or past research visualization of how concepts, constructs, vari-
studies. Generally, researchers build on the work ables, and operationalizations are related to one
of others and use existing concepts and defini- another in the research process. Sections on iden-
tions unless they are inadequate or inappropri- tifying and operationalizing variables are pre-
ate (Katzer et al., 1978). sented later in this chapter.
A concept, or the thing you want to study, It is important to note that both concepts and
represents a number of individual, but related, constructs are arbitrary creations of research-
things. It is an abstract way of thinking that helps ers (Kibler, 1970). As a result, in reading many
us group together those things that are similar to research reports about the same topic, you are
one another and, at the same time, distinguish likely to find considerable variation in how
them from dissimilar other things. A concept scholars describe concepts and constructs.
can be an object, an event, a relationship, or a Generally, researchers describe the theoretical
process. Examples of concepts include a faulty foundation of a study in the literature review.
argument, an effective public speaker, a con- This is where concepts and constructs are intro-
flict between spouses, leadership, underrepre- duced. Then, in the hypotheses and research
sentation of minorities on prime-time television questions, constructs are further defined as
shows, and so on. Even though they are intended variables. Operationalizations of each variable
to represent a class of things that have common are presented in the methods section of written
characteristics, a concept does not have a fixed research reports.
or precise meaning or definition (de Vaus, 2001). Variables are the elements of interest to
With respect to the research literature, concepts researchers. Finding new descriptions, explana-
are generally introduced early in the literature tions, or predictions for variables is the primary
review with a general description. motivator for conducting scholarly research. To
In some cases, a set of concepts can be con- better understand how variables are used in the
nected to form a conceptual scheme. For exam- research process, we need to consider hypoth-
ple, a researcher could identify ways in which eses and research questions.

key06767_ch03_035-056.indd 41 1/25/10 6:51:49 PM

 Rev.Confirming Pages

42 CHAPTER 3 / INTRODUCTION TO QUANTITATIVE RESEARCH

Concepts

violence

children TV

Constructs

children Concepts become

TV constructs when
linked together in
violence meaningful ways

Variables

H1 : Hypothesis
Children are influenced by violence on TV

time
of number
day of hours
home
school
children Operationalizations viewing
alone or
with
private parents
public school
school children
children
children
TV

dramatic
violence violence

video cartoon
game violence
violence

FIGURE 3.2 Moving from Concepts to Constructs to Variables to Operationalizations

key06767_ch03_035-056.indd 42 1/25/10 6:51:49 PM

 Rev.Confirming Pages

RESEARCH HYPOTHESES FOR QUANTITATIVE RESEARCH 43

RESEARCH HYPOTHESES FOR rejection (Young, Paxman, Koehring, & Ander-

QUANTITATIVE RESEARCH son, 2008, p. 58).
Hypothesis: Rejection messages from friends
Scholars rely on hypotheses to direct their will be perceived by would-be lovers as
quantitative inquiry. A hypothesis is an edu- more polite, pleasant, and appropriate than
cated guess, or a presumption, that is based on rejection messages from romantic partners and
a scholar’s review of the research literature. It acquaintances.
describes a logical explanation of the difference
or relationship between two or more variables. Directional Hypotheses
To be tested through research, the consequences
of the hypothesis must be observable. In fact, the Return to the hypothesis about the politeness
researcher designs the research study to test the of rejection messages. In this case the research-
relationships described in the hypothesis. ers expect to demonstrate that would-be lovers
For example, the following hypothesis will report that one type of relational partner
describes a relationship between romantic part- (friends) will be more polite than acquain-
ners’ anticipation of more online interaction and tances or romantic partners when they reject
their uncertainty about the relationship (Pauley the romantic request. Because the hypothesis
& Emmers-Sommer, 2007, p. 413). explicitly states which type of relational partner
Hypothesis: Anticipation of future interac- will be more polite, the hypothesis is directional.
tion with an online partner is associated with Thus, a directional hypothesis is a precise state-
greater reported uncertainty reduction in ment indicating the nature and direction of the
relationships. relationship or difference between the variables.
The hypothesis proposes what the relation-
ship will be. When those involved in online Nondirectional Hypotheses
romantic relationships have the expectation of A nondirectional hypothesis is less explicit. Con-
more interaction with their online relational tinuing with the rejection message example, a
partner, their uncertainty about that relationship nondirectional hypothesis could be
will be reduced. A hypothesis states the nature of
the relationship between variables. This is differ- Hypothesis: Would-be lovers will report dif-
ent from phrasing the relationship as a research ferences in the politeness of rejection messages
question, which would ask what the relationship based on relationship type (acquaintance,
would be. friend, romantic partner).
As another example, a research team exam- The nondirectional hypothesis states that a
ined differences in the politeness of rejection difference will occur but does not state the direc-
messages. Unrequited love is when one person tion of the difference. Acquaintance, friend, and
has romantic feelings for another person, but the romantic partner are three categories of the rela-
other person does not want to be romantically tionship type variable examined in the study. The
involved. The person with the romantic feelings nondirectional hypothesis indicates that there
is labeled as the would-be lover. In this study, will be a difference in the degree of politeness
research participants were asked to think about a among the three types of relational partners,
time when they told someone they were roman- but does not predict which type of partner will
tically interested in him or her and the other deliver a more polite rejection message. Some
person indicated that romantic interest was not researchers judge directional hypotheses to be
mutual. Participants reported on rejections from more sophisticated than nondirectional hypoth-
acquaintances, friends, or romantic partners. eses because they more explicitly state what
Based upon the literature, the research team differences will be found. Both types of hypoth-
wrote a hypothesis that predicted which type eses are found in the communication research
of relational partner will deliver the most polite literature.

key06767_ch03_035-056.indd 43 1/25/10 6:51:49 PM

 Rev.Confirming Pages

44 CHAPTER 3 / INTRODUCTION TO QUANTITATIVE RESEARCH

Assessing Hypotheses research hypothesis. A null hypothesis states that

no difference or no relationship, except one due
The following criteria can help you assess the to chance, exists between the variables. As you
clarity and utility of hypotheses (Hoover & can see, the null hypothesis is in direct opposi-
Donovan, 1995; Pyrczak & Bruce, 2007; Salkind, tion to the research hypothesis. According to the
2008). The first criterion asks evaluative ques- traditions of science, the null hypothesis must be
tions about wording, such as, Is the hypothesis assumed to be true until support for the research
simply stated? Is the hypothesis stated as a sin- hypothesis can be demonstrated.
gle declarative sentence? Are at least two vari- Although it is unlikely that you would see a
ables identified? null hypothesis presented in scholarly work, it is
The second criterion focuses on the variables helpful to think through what the null hypoth-
in the hypothesis. Are the variables clearly speci- esis would be. As an example, refer back to the
fied? Can the variables be measured by tech- research hypothesis about unrequited love rejec-
niques you know how to use? If not, you will tion messages. The research hypothesis was pre-
need to learn the appropriate measurement or sented as
observation method or ask someone to help you
with this aspect of the study. H: Rejection messages from friends will be per-
A third criterion for assessing hypotheses is ceived by would-be lovers as more polite, pleas-
to examine how the differences or relationships ant, and appropriate than rejection messages
between variables are stated. Are they stated from romantic partners and acquaintances.
precisely? A good way to test this is to explain The null hypothesis could be written as
a hypothesis to someone who is unfamiliar with H0: There will be no difference in would-be
research methods. If you can describe it to some- lovers’ perceptions of the politeness of rejec-
one else, the difference or relationship in the tion messages from acquaintances, friends, or
hypothesis is likely to be precise enough to use romantic partners.
in a research study. The fourth criterion asks if It could also be written in one of the following
the hypothesis reflects a theory or a body of liter- ways:
ature. In other words, is the hypothesis logically
supported by or based on previous research? H0: Would-be lovers will not report differ-
The fifth criterion asks if the hypothesis is ences in perceptions of the politeness of rejec-
testable. In other words, can the difference or tion messages delivered by acquaintances,
relationship between the variables be observed friends, or romantic partners.
or demonstrated in some way? This criterion is H0: Degree of perceptions of politeness in rejec-
crucial to conducting the research study. The tion messages reported by would-be lovers will
method, or the way in which the study is con- not vary by relational partner type (acquain-
ducted, must match the hypothesis. Evaluating tance, friend, romantic partner).
hypotheses is important because they guide
Notice that the symbol H indicates the
how the research project is developed and
research hypothesis and the symbol H0 indicates
structured.
the null hypothesis. In some journals, you may
also see the symbols H1 or Halt to indicate the
Null Hypotheses research hypothesis.
Although a scholar spends time developing
research hypotheses from the published litera- Research Traditions in the Use of Hypotheses
ture, these are not directly tested. Rather, a sta-
tistical test is performed on the null hypothesis, You should be aware of several traditions regard-
or the implicit complementary statement to the ing the use of hypotheses (Kibler, 1970). The first
research hypothesis. In using a statistical test, is that hypotheses are always tentative. Even
the researcher hopes to reject the null hypoth- when support for a hypothesis is found, sup-
esis and, in effect, validate the alternative, or the port is never considered absolute. Knowledge is

key06767_ch03_035-056.indd 44 1/25/10 6:51:49 PM

 Rev.Confirming Pages

TYPES OF VARIABLES 45

developed over time, and the continual efforts Another opportunity for researchers to use
of researchers contribute to the development research questions arises when several expla-
of theory as they seek to confirm or disconfirm nations could exist for the proposed relation-
hypotheses. Thus, one study cannot absolutely ship. For instance, Yoo’s (2009) literature review
confirm or disconfirm any relationship or differ- offered two competing evaluations for how
ence between or among variables. likeable someone is if they complain or share
Researchers honor the second tradition when negative information with a conversational part-
they present the research hypothesis. Although ner. One explanation is that sharing negative
not explicitly stated in a journal article, the null information can be perceived as a sign of bond-
hypothesis is assumed to be in direct opposition ing between two people. The other explanation
to the research hypothesis. The research hypoth- is that sharing negative information is much like
esis, the one the researcher wants to confirm, gossiping and previous research demonstrates
states the way in which the variables are differ- that when people share negative information
ent or the way in which the variables are related while gossiping, their likeability is decreased.
to one another. Alternatively, the null hypoth- Which is a better explanation? Yoo developed
esis, even when not given, implies that no dif- and tested this research question to find out:
ference exists between the variables or that the “Will participants who share only negative
variables are not related in any way. In nearly information about a target rate a greater liking of
every case, the research, or alternative, hypoth- their partner than participants who share posi-
esis is the focus of a research project. tive or combined information?” (p. 33).
A research question makes more sense in
this instance because previous research could
RESEARCH QUESTIONS IN not help the researcher make an educated and
QUANTITATIVE RESEARCH informed guess between the two explanations.
In both of these cases, the research question is an
Why would a researcher ask a research ques- opportunity for the researcher to describe com-
tion instead of stating an educated guess with munication phenomena. Research questions are
a hypothesis? There are several occasions when primarily tools of descriptive research.
asking a research question has more utility. The
first is when little is known about a communica- TYPES OF VARIABLES
tion phenomenon. For example, some commu-
nication phenomena—particularly those linked In hypotheses and research questions for quan-
to the use of media or technology—enter the titative studies, researchers make the object of
mainstream faster than researchers can study their study more explicit by further specifying a
their effects. As a result, Campbell (2008) had construct as a variable. A variable is an element
few research studies on which to base research that is specifically identified in the research
hypotheses when he explored how mobile phone hypotheses or questions. The literature review
use in public settings was influenced by cultural may describe several related concepts and con-
and individual differences. Although studies had structs, whereas the hypotheses and research
looked at mobile phone use and the perceptions questions identify a more limited set of variables
of users, there were few studies upon which to to be explored.
develop hypotheses regarding how individuals The study of communication from a social
interpret rules or norms for mobile phone use science perspective borrows heavily from the
in public settings. Thus, the following research traditional science disciplines. As a result, quan-
question was asked: titative research relies on the study of variables.
RQ: To what extent are individualism and In social science, variables are the properties or
collectivism related to assessments of mobile characteristics of people or things that vary in
phone use in public settings that support quality or magnitude from person to person or
different levels of focus? object to object (Miller & Nicholson, 1976).

key06767_ch03_035-056.indd 45 1/25/10 6:51:49 PM

 Rev.Confirming Pages

46 CHAPTER 3 / INTRODUCTION TO QUANTITATIVE RESEARCH

Simply, to be a variable, the element must impossible to see an attitude about violence on
vary. In other words, the variable must have two television. But a researcher can see the behav-
or more levels. For example, sex is a variable as ioral effects of your attitude when you report
it varies between male and female. Leadership liking shows with violent content or when you
can be a variable, as leaders can be described as change the channel to avoid watching such
autocratic or democratic. Communication appre- programming.
hension, like other constructs measured by ques- Whether they are tangible or constructed,
tionnaires, is a variable: Individuals’ scores can when they are included in research questions or
have many levels, ranging between a potential hypotheses, variables must also be identified as
minimum and maximum score. independent or dependent. Simply, the indepen-
Communication phenomena or communica- dent variable is presumed to have an effect on
tion-related phenomena do not always take on or cause a change in the dependent variable. The
the role of variables in communication research. sections that follow describe how the two types
For example, even though sex varies, it cannot of variable are used in research and explain the
be a variable if the researcher studies only the nature of the relationships between and among
leadership abilities of women. Leadership ability variables. In quantitative research, the researcher
would be a variable; sex would not. In another should specify which variables are independent
example, leadership style cannot be a variable and which are dependent.
if the researcher examines only ways in which
democratic leaders encourage group members to Independent Variables
participate. In this case, encouragement of group
member participation would be a variable; dem- Variables manipulated by the researcher are called
ocratic leadership style would not. Something independent variables. Presumably, the manipu-
can be identified as a variable only if it fluctu- lation, or variation, of this variable is the cause of
ates in the research study (Kibler, 1970). This change in other variables. Technically, the term
can be confusing because not every concept that independent variable is reserved for experimental
can vary and act as a variable does so in every studies (see Chapter 8) in which the researcher has
research study. control over its manipulation. In some research
Some variables are easy to view and count. reports, independent variables are referred to as
They are tangible and observable. For example, antecedent variables, experimental variables,
it is easy to identify and count what type and treatment variables, and causal variables.
how many hand gestures a person makes while The independent variable is manipulated
arguing. Alternatively, it may not be possible because the researcher wants to assess if differ-
to directly view some variables. In these cases, ent types or values of the variable result in some
researchers develop constructs to represent relationship with another observed phenom-
properties that control other events. For exam- enon, or the dependent variable. If the effect or
ple, communication competence is a construct relationship occurs, the researcher presumes that
that was developed by communication scholars the independent variable caused the change in
to represent perceptions of effectiveness and the dependent variable.
appropriateness of interaction. There is indirect For example, Horan and his research team
evidence that this construct exists (after all, most (2009) wanted to examine how illnesses that are
people have ideas about what constitutes com- generally invisible to others would influence the
munication competence), but the construct itself type and amount of communication from a con-
is invisible. Communication researchers will- versational partner. Participants in the study read
ingly accept constructs because they are able to a hypothetical scenario about a new neighbor.
demonstrate differing effects produced by vari- There were three scenarios. In one scenario, the
ables. For example, competent communicators target had cancer. In a second scenario, the target
produce different effects on their receivers than was HIV+. In the third condition, the target was
incompetent communicators do. Similarly, it is healthy. Use of the three scenarios was meant to

key06767_ch03_035-056.indd 46 1/25/10 6:51:49 PM

 Rev.Confirming Pages

TYPES OF VARIABLES 47

DESIGN What Is Sex? What Is Gender?

CHECK
Many quantitative communication studies use sex and gender interchangeably as
an independent variable. Unfortunately, many authors do not give their opera-
tionalization of this variable. When sex or gender is not operationalized, it is prob-
ably safe to assume that the researcher had participants self-report their biological
sex—male or female. Of course, biological sex is not the same as one’s gender or
psychological orientation to sex—for example, feminine, masculine, or androgy-
nous. There is a difference in being female and being feminine. This confusion has
become even greater because some editors and publishers have requested that the
more politically correct term gender be substituted for the more traditional term
sex. If sex or gender is not operationalized, read the methods section carefully to
determine if participants self-reported this information, if researchers assumed this
information about participants, or if participants responded to a measuring instru-
ment. If in doubt, use the terms as they are used by the authors of the published
research reports.

manipulate the independent variable of illness to independent variables in their statistical tests.
see how research participants would communi- Obviously, the research team could not alter or
cate with their hypothetical new neighbor. change these aspects of the study’s participants.
Traditionally, the independent variable is Thus, these two variables varied naturally, and
manipulated in the context of an experiment to the research team used information participants
produce a change in the dependent variable. If provided to code them as being from one of four
all communication research were conducted as world regions (i.e., Americas, Asia, Europe, or
experiments, we could end our explanation of Middle East) and as female or male. Thus, both
independent variables here. However, commu- country of origin and sex had two or more levels
nication researchers use experimental, quasi- and could be used as predictor variables.
experimental, and descriptive research designs In a study of how people maintain romantic
(each of these is explained in detail in Chapter 8). relationships (Dainton, 2003), relational equity
Whereas experimental research designs use re- was used as an independent variable. During
searcher manipulation of the independent vari- the data collection procedures, the participants
able, other quantitative research designs do not. responded to questionnaire items that allowed
In these cases, the researcher relies on natural the researcher to determine if participants per-
variation in the independent variable to produce ceived they were in equitable romantic relation-
different effects on the dependent variable. In ships, or in romantic relationships in which they
those instances, the term predictor variable is were benefiting more or less than their partner.
preferred and should be used in descriptive re- Thus, the researcher used the natural variation
search designs (for example, survey research). in participants’ reports of their relational equity
Researchers are still interested in the predictor to create this predictor variable for the study.
variable’s effect on other variables, but distin- As you can see, it is better to describe the inde-
guishing between the two terms helps remind us pendent variable as the variable that alters or
of this key difference. A few examples will make changes the dependent variable. The researcher
this distinction clearer. does not always manipulate it. In many cases,
In a study of how immigrants are social- variation in the independent variable occurs
ized into American culture, Erbert, Perez, and naturally as a characteristic of the population
Gareis (2003) used country of origin and sex as under study. Any change or difference in the

key06767_ch03_035-056.indd 47 1/25/10 6:51:49 PM

 Rev.Confirming Pages

48 CHAPTER 3 / INTRODUCTION TO QUANTITATIVE RESEARCH

independent variable is the presumed cause of stake represented in each media sector. These
a change in the dependent variable if an experi- structural differences were expected to have
mental method is used. Causality is weakened if directional effects upon the networks’ ticker
the researcher does not directly manipulate the content” (p. 897). Researchers measured the
independent variable, but it would be impracti- dependent variable, ticker content, by counting
cal to conduct all communication research from the instances in which overt messages, such as
this experimental framework. Generally speak- “Go to cnn.com for the latest developments,”
ing, however, the independent variable is the appeared in 15 hours of programming. As sug-
variable a researcher predicts from, whereas the gested by the hypothesis, the research team
dependent variable is the one the researcher pre- wanted to see if overt promotional messages
dicts to (Kibler, 1970). were used by these news networks. Indeed, they
found that CNN, and to a smaller extent, Fox,
Dependent Variables used their news crawl space for self-promotion.
Thus, the dependent variable is the variable the
Even though the independent, or predictor, vari- researcher is trying to explain in a research proj-
able causes the change, the dependent variable ect (Kibler, 1970).
is of primary interest to the researcher, who is
trying to describe, explain, predict, or control
changes in it. The dependent variable is influ- The Relationship Between Independent
enced or changed by the independent variable. and Dependent Variables
Sometimes in descriptive research designs, the
In quantitative research, the dependent variable
terms criterion variable and outcome variable
is selected first because it is the primary topic
identify the dependent variable. Regardless of
of the research study (Thompson, 2006). Reread
what it is called, logically a researcher cannot
the hypothesis just presented. Ticker content, or
have a dependent variable without an indepen-
what information is presented on news tickers, is
dent variable, and vice versa.
the primary research topic. Thus, ticker content
Changes in the dependent variable are a con-
is the dependent variable. Then, logically, the
sequence of changes in the values or levels of the
researchers start to investigate what indepen-
independent variable. For example, in a study of
dent variables would influence or cause change
what type of content appears in the news tickers
in the dependent variable. When a researcher is
(or crawlers) of cable news network (CNN, FOX
looking to describe, predict, or explain differ-
News Channel, MSNBC), one hypothesis stated
ences or relationships, a hypothesis must have at
(Coffey & Cleary, 2008, p. 897):
least one independent and one dependent vari-
H1: Cable news networks use their news tickers able. Variables are identified as independent and
for overt promotional purposes. dependent to establish the presumed relation-
In this case, the independent variable appears ship between them.
to be cable news network with the CNN, FOX How many independent and dependent vari-
News Channel, and MSNBC as different catego- ables can a researcher specify? That depends on
ries. But close reading of the methods section of the nature or complexity of what the researcher
this research article reveals that the difference wants to explain. Most important, however, the
in the three categories of the independent vari- number of independent and dependent vari-
able is really based on the parent company’s ables dictates which statistical test will be used.
ownership structure (i.e., degree of integration, We will review these distinctions in Chapters 11
percent of revenue contribute to total com- and 12 as the most common statistical tests are
pany revenue). As the researchers explained: described and explained. But, for now, analyz-
“While all three cable networks are owned by ing a hypothesis from a research report will
horizontally and vertically integrated media help you understand the nature of the rela-
companies, each of these parent companies dif- tionship between independent and dependent
fers in terms of its holdings and the percentage variables.

key06767_ch03_035-056.indd 48 1/25/10 6:51:49 PM

 Rev.Confirming Pages

TYPES OF VARIABLES 49

Sparks, Pellechia, and Irvine (1998) wanted The use of “high credibility” could imply a
to study how television news affects viewers’ comparison to low credibility. But the research-
beliefs about UFOs (unidentified flying objects). ers do not offer an alternative to “high credibil-
They argued that although many people believe ity.” Thus, it does not appear that the researchers
that the mass media play a role in misleading are using the credibility of news reports as a
people so that they accept paranormal events variable. This is also apparent in the author’s use
like UFOs, little research has provided evidence of “supports the existence of UFOs.”
to substantiate that claim. Thus, this research Reread the hypothesis. The authors explicitly
team designed an experiment to test the possible state that viewers of the one-sided news report
impact of news reports about UFOs on subse- will express greater belief in UFOs. Also notice
quent UFO beliefs. Their hypothesis was their use of the word “than.” This word usage
H: Subjects who view a high credibility, and sentence construction implies that the view-
one-sided news report that supports the ers of the one-sided news report will differ from
existence of UFOs will, subsequently, express viewers of the two-sided news report. Thus,
greater belief in the existence of UFOs than will the authors have implied a variable—belief
subjects who view a two-sided news report in UFOs.
from the same program. (p. 287) The second type of comparison is when differ-
As a student, of course, you hope that the ences are explicitly stated. The authors make a clear
authors of the research report clearly identify and explicit distinction between one-sided and
which variables are independent and which two-sided news reports. So type of news report
variables are dependent. This is not always the (one-sided or two-sided) must be a variable.
case. Even if it is, you should be able to read Which variable is the independent variable
the research article and verify the independent and which variable is the dependent variable?
and dependent variables. One way to do so is to We could assume from the way the hypothesis is
look for what changes or varies in the hypoth- written that the authors use type of news report
esis statement. Examine Table 3.3. By comparing as the independent variable because they sus-
each mention of each element, we can determine pect that the variation in this element will cause
the variables and then assign them as indepen- a change in viewers’ beliefs in UFOs, the depen-
dent or dependent. dent variable. However, it is always a good idea
The authors made no distinction between the to read the methodology section of the research
two mentions of “subjects” in the hypothesis, so report to confirm your identification of variables
“subjects” is not a variable. Likewise, there is no and assignment of variables as independent and
distinction in the use of “view.” Thus, viewing is dependent, and to see exactly how the research-
not a variable. ers measured or observed their variables.

Table 3.3 Determining Variables from the Elements of a Hypothesis

First Mention of Element Second Mention of Element Type of Comparison Type of Variable
Subjects Subjects No distinction —
View View No distinction —
High credibility — No alternative given —
One-sided news report Two-sided news report Explicit comparison Independent variable
Supports the existence — No alternative given —
of UFOs
Greater belief in UFOs (Less belief in UFOs) Implied comparison Dependent variable

key06767_ch03_035-056.indd 49 1/25/10 6:51:49 PM

 Rev.Confirming Pages

50 CHAPTER 3 / INTRODUCTION TO QUANTITATIVE RESEARCH

OPERATIONALIZING VARIABLES A good example of the specificity needed for

operationalizations is demonstrated in a study
Whether variables are used as independent or that examined the influence of message design
dependent variables in hypotheses or questions, strategies on changes in attitude and use of mari-
researchers must take one additional step to actu- juana (Harrington et al., 2003). The first hypoth-
ally use the variables in a quantitative study. Each esis of the study stated:
variable must be operationalized, or observed or H1: Compared to low sensation value anti-
measured in a specified way. Most variables can marijuana messages, high sensation value
be operationalized in multiple ways. That is why anti-marijuana messages will lead to greater
an operationalization specifies the concrete steps attitude, behavioral intention, and behavior
or processes for creating and measuring the vari- change. (p. 22)
able. The operationalization is specific enough
that others reading the description of the opera- Although the research team identifies the
tionalization in the research report can follow it variables in the hypothesis, there is not enough
(Miller & Nicholson, 1976). information to replicate the research study. What
Researchers use three criteria for select- are low sensation value and high sensation value
ing among available operationalizations. First, messages? How do you measure attitude, behav-
which operationalization is practical and useful ioral intention, and behavior change?
in their study? Second, can a justifiable argu- In the methods section of their study, the
ment be developed for this operationalization? research team specifically defined, or opera-
Third, does the operationalization selected coin- tionalized, these variables. To create the high
cide with the conceptual definition? sensation and low sensation value messages,

TRY THIS! Identifying Independent and Dependent Variables

Following the process shown in Table 3.3, identify the independent and dependent
variables for each hypothesis in the following table.

Hypothesis Independent Variable Dependent Variable

Family conversation orientation is Family Adult children’s
positively related to adult children’s conversation emotional
emotional intelligence. (from Keaten orientation intelligence
& Kelly, 2008)
There will be a negative effect of per-
ceived peer norms for thinness on the
body esteem of young women. (from
Krcmar, Giles, & Helme, 2008)
Indirect forgiveness will be more
frequently used by friends than by
dating partners. (from Merolla, 2008)
There is a linear relationship between
the number of friends a profile owner
has and observers’ perceptions of
extraversion. (Tong, Van Der Heide,
Langwell, & Walther, 2008)

key06767_ch03_035-056.indd 50 1/25/10 6:51:49 PM

 Rev.Confirming Pages

MAKING THE CASE FOR QUANTITATIVE RESEARCH 51

the research team reviewed public service anno- article, researchers will be precise and provide
uncements (PSAs) currently being shown on the operationalization—how the variable will be
television. Identifying those that were high and observed and measured—of each variable.
low sensation, the research team then used focus Both researchers and consumers of research
groups to verify their characterizations. With benefit when there is specificity about what is
this information, the team created high sensation being studied (Miller & Nicholson, 1976). If terms
value PSAs that featured high intensity, loud are defined explicitly in operationalizations, it is
and driving music, quick and multiple edits, more difficult for the researcher to draw con-
unusual camera angles, and extreme close-ups. clusions beyond the boundaries established.
Alternatively, low sensation value PSAs featured Another advantage is that findings of studies
slower paced music, fewer edits, more typical that use the same or similar operationalizations
camera angles, and no extreme close-ups. Thus, can be compared. When an operationalization
the authors clearly described what constitutes a is accepted as a standardized way of measuring
high sensation and a low sensation value mes- or observing, there are two consequences. First,
sage and explained specifically how those mes- preciseness is achieved, which allows our under-
sages were created. standing of the communication phenomenon to
The researchers also explained how they oper- be enhanced, and this enhancement supports
ationalized each of the other three variables— theory development. Second, operationalizations
attitude change, behavioral intention change, and allow other researchers to replicate the research
behavior change. Attitudes toward marijuana use to verify or challenge findings.
were measured with questionnaire items from a
study done as an evaluation of a national drug
abuse campaign. Behavioral intention toward MAKING THE CASE FOR
using marijuana was measured with this and QUANTITATIVE RESEARCH
similar items: “How likely is it that you will use
in the next 30 days?” Marijuana use behav- For many of the other social sciences and for a
ior was measured with items that asked if par- long period of time in communication, research-
ticipants had used marijuana in the past 30 days, ers conducting their research from the social sci-
the past year, or at any time during their lifetime. ence perspective relied primarily on quantitative
For each variable the research team provided research methods. Traditionally, experimental
examples of the items and indicated the direc- forms were preferred to other methods in which
tion of the desired change (i.e., should watching variables are quantified (for example, surveys).
a PSA result in a higher or lower score). Today, however, communication researchers
Thus, different terms are used as the researcher choose from a variety of quantitative and qualita-
moves through conceptual, theoretical, and tive methods. Which method should be chosen
empirical stages. Occasionally, you will notice is one of the decisions the researcher must make.
that the terms are used interchangeably. But, tech- By answering the question “What do I want to
nically, as the terms move from concept through know in this study?” clearly and completely, the
construct through variable to operationalization, researcher can select the appropriate methodol-
a greater level of specificity should be given. ogy. Reviewing the advantages and limitations
When you read scholarly journal articles of quantitative research will help you determine
using quantitative methods, you will read about what types of questions are best answered by
concepts and constructs in the literature review. quantitative methods.
In this section, the researchers may even present
conflicting definitions of the concept of interest. Advantages of Quantitative Research
When researchers present their research ques-
tions and hypotheses, they will become more The advantages of using quantitative research
specific and treat the concepts and constructs as methods are obvious. First, quantitative meth-
variables. In the methods section of the journal ods used in communication research follow the

key06767_ch03_035-056.indd 51 1/25/10 6:51:49 PM

 Rev.Confirming Pages

52 CHAPTER 3 / INTRODUCTION TO QUANTITATIVE RESEARCH

TRY THIS! Operationalizing Variables

Communication researchers are often interested in studying concepts that are
abstract in nature. For example, to study how couples in love communicate and
then compare that to how couples not in love communicate, we would first have to
describe what love is and then operationalize it. Review the examples that follow;
which operationalization do you like best? What is problematic with the two opera-
tionalizations? What operationalizations can you develop? Are there other descrip-
tions of love more central to communication?
Description of Love Operationalization of Love
When two people hold hands in The number of times a couple hold hands
front of others for longer than one minute in public
When two people tell their friends The degree of vocal intensity in telling
and relatives that they are in love friends and relatives that they love the
other person; the number of reasons given
when telling friends and family that they
love the other person.
When two people independently
feel attracted to each other and
decide not to date others
When two people in a committed rela-
tionship do unexpected tasks for each
other

tradition and history of quantitative methods and relationships, researchers are able to be
in other disciplines. Thus, the use of quantita- very precise and exact in their comparisons. This
tive methods implies a certain amount of rigor is especially important in the study of micro-
in the research process. By quantifying and mea- elements of communication. Quantifying abstract
suring communication phenomena, communi- concepts provides researchers a way to isolate
cation researchers are using the same research variables and gain knowledge about concepts
language as researchers with whom they share that would otherwise remain hidden.
interests. Sharing a research tradition could, for Third, because we can quantify communica-
example, strengthen the relationship between tion phenomena, we can make comparisons, and
communication researchers who study organiza- those comparisons can be made among a large
tional communication and researchers from the group of participants. As a result, researchers
management discipline. Likewise, many com- can generalize their findings to other individuals
munication researchers who study interpersonal who have the same characteristics as those in the
issues would share a research tradition with psy- research project.
chologists who focus on individuals and their
relationships. Limitations of Quantitative Research
A second advantage of quantitative research
comes from the use of numbers and statistics. By Of course, with advantages come limitations.
quantifying communication concepts and using And, as is often true, limitations stem from
statistical procedures for evaluating differences the same sources as the advantages. Because

key06767_ch03_035-056.indd 52 1/25/10 6:51:49 PM

 Rev.Confirming Pages

ISSUES OF RELIABILITY AND VALIDITY 53

quantitative research can focus on micro- reliability and validity in the planning stages of
elements of communication phenomena, this their research project. But doing so is not enough.
type of research generally does not lend itself to At the end of data collection, one of the first ana-
capturing the complexity or depth of communi- lytical steps quantitative researchers perform is
cation over time. The restriction of focusing on to assess the reliability and validity of their data.
just a few variables at a time makes it more dif- Without reliable and valid measurement, the
ficult for researchers to examine the entirety of results of the study, which rely on how the data
the communication process. are collected, are suspect. More detailed infor-
Likewise, quantitative research cannot cap- mation on how to assess reliability and validity
ture communication phenomena that cannot be can be found in Chapter 6.
replicated or simulated in a controlled research
environment. Although researchers can use Threats to Reliability and Validity
quantitative methods in the field, all participants
must be subjected to the same stimuli and sys- Any measuring device can be faulty—whether
tematic procedures. As a result, questions about the device is a questionnaire or a researcher cat-
a communication phenomenon that occurs spon- egorizing and counting the number of times you
taneously or sporadically are not as well suited use “Ms. Wright” while talking to your boss.
to quantitative methods. Communication research is especially vulnera-
ble to threats to reliability and validity because it
measures complex communication phenomena.
ISSUES OF RELIABILITY AND VALIDITY Let’s examine different types of threats to reli-
ability and validity (Selltiz et al., 1959).
All quantitative research hopes to be both reliable First, reliability and validity are threatened if
and valid. Reliability is achieved when research- the measuring device cannot make fine distinc-
ers are consistent in their use of data collection tions. To what degree can the measuring device
procedures and when participants react simi- capture specificity or exactness? In measuring a
larly to them. Reliability also means that other person’s orientation, or relationship, to organiza-
researchers using the same measure in another tions, can the measure of organizational orienta-
project with comparable participants would get tion (McCroskey, Richmond, Johnson, & Smith,
similar results (Hoover & Donovan, 1995). 2004) really distinguish among individuals who
But reliability is only part of evaluating a quan- have a strong affinity for organizations, those
titative research method. Validity is achieved who do not care much about organizations, and
when the measurement does what it is intended to those who cannot adapt to organizations well?
do (Hoover & Donovan, 1995). Validity is related Second, reliability and validity are threatened if
to truth. Thus, within scientific reasoning, the bet- the measuring device cannot capture how people
ter the technique is at uncovering the reality of differ. Not only must the organizational orienta-
the concept, the more valid it is. If a questionnaire tion measure make distinctions among the three
presumably testing your degree of communica- different orientations a person can have to orga-
tion competence asks you to respond to an item nizations, but it must also provide information
about your ability to handle conflict, the item about how individuals with those orientations
would be valid if you and other people in gen- communicate in organizations.
eral believe that handling conflict effectively is, A third threat is present when researchers
in fact, evidence of communication competence. attempt to measure something that is irrelevant
Reliability and validity are related. Both are or unknown to the respondent. Anyone can
evaluations of the utility of a measuring device. respond to a questionnaire; but if the respon-
When reliability and validity are achieved, the dent does not understand the questions or if the
data are free from systematic errors (Selltiz, respondent is asked to give responses for some-
Jahoda, Deutsch, & Cook, 1959). This is why thing for which he or she has no reference, then
quantitative researchers must carefully consider the measurement is faulty. This is especially true

key06767_ch03_035-056.indd 53 1/25/10 6:51:49 PM

 Rev.Confirming Pages

54 CHAPTER 3 / INTRODUCTION TO QUANTITATIVE RESEARCH

of opinions and attitudes, which are often part of if personality is the best choice of independent
quantitative communication research. Simply, it variable. Rather than personality, could it be
would not make sense to ask an unmarried and that our experiences with conflict help mold our
never-married respondent to answer questions preferences?
about communication satisfaction with a marital 2. Variation or differences due to personal
partner. factors, such as mood, fatigue, health, time of
Finally, the complexity of human communica- day, and so on. How would these factors affect
tion behavior can threaten reliability and valid- your choice of conflict management strategy?
ity. Can any measuring instrument really capture
3. Variation or differences due to situational
the phenomenon? For example, if two research-
factors. Would your choice of conflict manage-
ers are measuring the length of a building and
ment strategy be different at school, at work, at
they disagree, together they can remeasure the
home?
building. Their individual measures can be veri-
fied by each other or a third party. But how can 4. Variation due to differences in how the
someone verify your organizational orientation? research project is administered. Different
Asking you to respond again to the questionnaire researchers may use different communication
may produce different results, depending on the styles in working with research participants.
number and types of interactions you have had Would you respond differently to a researcher
in organizations since you last answered the who appears to be bored than you would respond
questionnaire or depending on a change in your to a researcher who appears enthusiastic?
motivation or employment status. Someone else 5. Variation due to the number of items
cannot independently verify your organizational included in the measuring device. Asking only
orientation—only you can provide that infor- one question about each of the conflict manage-
mation. Moreover, what if your organizational ment styles would not be the same as asking sev-
orientation is not represented by the three orien- eral questions about each style. Multiple items
tations captured by this questionnaire? are required to capture the breadth and com-
As you can see, quantifying communication plexity of most communication phenomena.
phenomena provides some difficulties and cre- 6. Variation due to unclear measuring device.
ates consequences for the validity and reliability As an assistant to a researcher, you are asked
of the research. Although researchers want to to observe students giving speeches and count
capture variation across the participants in the the number of times students use incomplete
research project, the variation should be rep- sentences. In one speech, the speaker gives an
resentative of the true differences among indi- incomplete sentence but before moving on to the
viduals and not result from measurement errors next sentence catches the error and restates the
(Selltiz et al., 1959). incomplete sentence as a full one. Do you count
Besides the variation that researchers want to this or not?
capture, there are many other possible sources of
variation (Selltiz et al., 1959). Unless these varia- 7. Variation affected by mechanical or pro-
tions are intended to be examined as part of the cedural issues. In a research experiment, you
research design, when they are present, the fol- are asked to recall your last conflict with your
lowing must be considered as threats to reliabil- relational partner and write out what you said
ity and validity: to him or her. As you recall the conflict incident,
you recognize that the conflict was a lengthy one
1. Variation due to factors not measured in lasting for about 20 minutes. But the researcher
the research study. For example, you want to left only 3 inches of space on the questionnaire
measure effects of personality on conflict man- for you to describe what you said. Not knowing
agement strategy choice. A researcher would what to do, you select the most important things
have to check the research literature to determine you said and leave out the rest.

key06767_ch03_035-056.indd 54 1/25/10 6:51:49 PM

 Rev.Confirming Pages

SUMMARY 55

8. Variation due to the statistical processing from concepts to constructs to variables to

of the data. For example, the researcher acciden- operationalizations in providing the degree
tally selects the wrong statistical test from the of objective specificity needed to examine
pull down menu. Or the person who enters your communication phenomena.
data in the computer pushes “4” every time you 7. Operationalizations are the specific way
have responded with a “5.” in which researchers observe and measure
Obviously, you want to capture and test vari- variables in quantitative research.
ation in your research project. But this type of 8. Quantitative research typically relies on
variation is planned for and identified in your the use of hypotheses to drive the research
hypotheses and research questions. Other varia- process.
tion is problematic. Look again at the model for 9. Hypotheses should be simply stated, have
quantitative research in Figure 3.1. Notice that variables and their relationships clearly
selection of methods must be balanced with specified, and be testable.
issues of reliability and validity. That is why
researchers using quantitative methods must 10. Although researchers develop research
carefully consider and evaluate their research hypotheses, the null hypothesis is actually
design, or their procedures and methods, before the focus of the statistical test.
data are collected. Capturing variation beyond 11. A hypothesis includes both independent
the variation the researchers want to capture is and dependent variables.
measurement error, which produces threats to 12. Researchers can also use research ques-
reliability and validity. tions as a foundation for their quantitative
research. Research questions are appropri-
ate to use when there is little known about
SUMMARY a communication phenomenon or when
previous results are inconclusive.
1. Quantitative research relies on the use of 13. Advantages of quantitative research
numbers as a way of observing and measur- include a certain degree of rigor, objectivity
ing communication phenomena. achieved through the use of numbers and
2. Researchers bring objectivity to the study statistics, and ability to make comparisons
of communication through the use of tradi- among a large group of participants.
tional quantitative approaches and statisti- 14. Limitations of quantitative research
cal techniques. include difficulty in capturing the com-
3. Quantitative research relies on deductive plexity or depth of communication over
reasoning. time, and the inability to capture com-
4. The primary objective of quantitative munication phenomenon that cannot be
research is to test propositions developed replicated or simulated in a controlled
from theory. environment.
5. The quantitative research model includes 15. Quantitative research must address threats
five components: research purpose, litera- to reliability and validity, including using
ture foundation, research questions and imprecise measures of variables, attempt-
research hypotheses, research methods, and ing to measure something that is unknown
validity and reliability. or irrelevant to participants, and difficulty
in capturing the complexity of human
6. Quantitative research requires that every interaction.
phenomenon studied be conceptualized and
then explicitly defined. Researchers work

key06767_ch03_035-056.indd 55 1/25/10 6:51:50 PM

 Rev.Confirming Pages

56 CHAPTER 3 / INTRODUCTION TO QUANTITATIVE RESEARCH

KEY TERMS See the website www.mhhe.com/keyton3 that

accompanies this chapter. The site contains
antecedent variable nondirectional a chapter summary, outline, and checklist. A
hypothesis PowerPoint presentation created by Dr. Keyton
causal variable
is also available for download, as well as a
concept null hypothesis
short multiple-choice quiz to check your learn-
conceptual scheme operationalization ing. You will also find a list of Internet resourc-
construct outcome variable es that provide further information about the
predictor variable issues in this chapter.
criterion variable
dependent variable reliability
directional hypothesis treatment variable
experimental variable validity
independent variable variable

key06767_ch03_035-056.indd 56 1/25/10 6:51:50 PM