Research Methods in Physical Activity 7th Edition Edition Jerry R. Thomas 2024 Scribd Download
Research Methods in Physical Activity 7th Edition Edition Jerry R. Thomas 2024 Scribd Download
com
https://ebookname.com/product/research-methods-in-physical-
activity-7th-edition-edition-jerry-r-thomas/
OR CLICK BUTTON
DOWLOAD NOW
https://ebookname.com/product/field-methods-in-archaeology-7th-
edition-thomas-r-hester/
https://ebookname.com/product/financial-accounting-7th-edition-
jerry-j-weygandt/
https://ebookname.com/product/nutrition-physical-activity-and-
health-in-early-life-2nd-edition-jana-parizkova-author/
https://ebookname.com/product/research-methods-for-social-
work-7th-ed-international-ed-edition-babbie/
Physical Activity and Health 1 Reprint Edition Nicholas
G. Norgan
https://ebookname.com/product/physical-activity-and-
health-1-reprint-edition-nicholas-g-norgan/
https://ebookname.com/product/business-research-methods-11-ed-
edition-donald-r-cooper/
https://ebookname.com/product/research-methods-for-criminal-
justice-and-criminology-7th-edition-michael-g-maxfield/
https://ebookname.com/product/introduction-to-kinesiology-
studying-physical-activity-3rd-edition-shirl-j-hoffman/
https://ebookname.com/product/cliffstestprep-act-cliffs-test-
prep-act-7th-edition-jerry/
2
Research Methods in Physical Activity
Seventh Edition
Human Kinetics
3
Library of Congress Cataloging-in-Publication Data
Thomas, Jerry R., author.
Research methods in physical activity / Jerry R. Thomas, Jack K. Nelson, Stephen J. Silverman. -- Seventh edition.
p. ; cm.
Includes bibliographical references and indexes.
I. Nelson, Jack K., author. II. Silverman, Stephen J., author. III. Title.
[DNLM: 1. Motor Activity. 2. Kinesiology, Applied. 3. Physical Education and Training. 4. Research Design. 5. Sports Medicine--methods. WE 103]
GV361
613.7'1072--dc23
2014028466
ISBN: 978-1-4504-7044-5 (print)
Copyright © 2015, 2011, 2005 by Jerry R. Thomas, Jack K. Nelson, and Stephen J. Silverman
Copyright © 2001, 1996, 1990, 1985 by Jerry R. Thomas and Jack K. Nelson
All rights reserved. Except for use in a review, the reproduction or utilization of this work in any form or by any electronic, mechanical, or other means,
now known or hereafter invented, including xerography, photocopying, and recording, and in any information storage and retrieval system, is forbidden
without the written permission of the publisher.
The web addresses cited in this text were current as of February 2015, unless otherwise noted.
Acquisitions Editor: Amy N. Tocco
Developmental Editor: Melissa J. Zavala
Managing Editor: Carly S. O’Connor
Copyeditor: Patsy Fortney
Indexer: Nancy Ball
Permissions Manager: Dalene Reeder
Senior Graphic Designer: Keri Evans
Cover Designer: Keith Blomberg
Photo Asset Manager: Laura Fitch
Visual Production Assistant: Joyce Brumfield
Photo Production Manager: Jason Allen
Art Manager: Kelly Hendren
Associate Art Manager: Alan L. Wilborn
Illustrations: © Human Kinetics, unless otherwise noted
Printer: Walsworth Print
Printed in the United States of America
10 9 8 7 6 5 4 3 2 1
Human Kinetics
Website: www.HumanKinetics.com
United States: Human Kinetics
P.O. Box 5076
Champaign, IL 61825-5076
800-747-4457
e-mail: [email protected]
Canada: Human Kinetics
475 Devonshire Road Unit 100
Windsor, ON N8Y 2L5
800-465-7301 (in Canada only)
e-mail: [email protected]
Europe: Human Kinetics
107 Bradford Road
Stanningley
Leeds LS28 6AT, United Kingdom
+44 (0) 113 255 5665
e-mail: [email protected]
4
Australia: Human Kinetics
57A Price Avenue
Lower Mitcham, South Australia 5062
08 8372 0999
e-mail: [email protected]
New Zealand: Human Kinetics
P.O. Box 80
Mitcham Shopping Centre, South Australia 5062
0800 222 062
e-mail: [email protected]
E6201
5
Contents
Preface
Study Tips
Acknowledgments
6
Describing Design and Analysis
Establishing Cause and Effect
Interaction of Participants, Measurements, and Treatments
Summary
7
Multivariate Forms of Correlation
Summary
David K. Wiggins
Daniel S. Mason
Development of the Discipline
Theory and Sport History
Relationship Between Theory and Method
8
Research Sources
Research Topics
Research Design
Data Analysis and Interpretation
Research Findings
Exemplary Studies in Sport History
Summary
R. Scott Kretchmar
Tim Elcombe
Identifying the Purposes of Philosophical Research
Identifying the Range of Philosophical Research
Locating a Research Problem
Analyzing a Research Problem
Summary
Barbara E. Ainsworth
Charles E. Matthews
Observational Versus Experimental Research
9
What Is Epidemiology?
Physical Activity Measurement Definitions
Assessment of Physical Activity
Epidemiological Study Designs
Reading and Interpreting a Physical Activity Epidemiological Study
Summary
10
Handling Multiple Experiments in a Single Report
Using Tables and Figures
Summary
Appendix
Statistical Tables
References
11
Preface
With the publication of this seventh edition, 30 years have passed since the first edition was
published in 1985. We take this opportunity to thank all the people who have used this
book over the years. We hope that you have learned a lot about research methods in the
study of physical activity. Maybe you have even enjoyed the humorous stories, jokes, and
pictures that we have included to enliven the reading. We were particularly pleased to note
the paper by Silverman and Keating (2002) in Research Quarterly for Exercise and Sport that
indicated that 71% of the people responding to their survey about the research methods
course used our book. Their survey results clearly show that the content of this book is well
aligned with the topics that teachers of research methods believe are important. We are also
delighted that many of you in other English-speaking countries have also used this book
—Good on ya, mates! In addition, we appreciate that earlier editions have been translated
into Chinese (twice), Greek, Korean, Italian, Japanese, Spanish, and Portuguese.
Dr. Jack Nelson has retired but continues to provide assistance and advice. Dr. Stephen
Silverman joined us as a coauthor on the fifth edition and, in spite of our sense of humor,
agreed to continue on the sixth and seventh editions. Dr. Silverman is a well-known scholar
and methodologist in physical education pedagogy and is a former editor in chief of
Research Quarterly for Exercise and Sport.
The main use of this text still appears to be in the first graduate-level research methods
courses, although it is also being used in undergraduate research methods courses and as a
resource for those engaged in research planning and analysis. Our use of the term physical
activity in the book title is meant to convey the broadly conceived field of study often
labeled kinesiology, exercise science, exercise and sport science, human movement, sport
studies, or physical education, as well as related fields such as physical therapy,
rehabilitation, and occupational therapy. We hope that everyone who reads, understands,
plans, carries out, writes, or presents research will find the book a useful tool to enhance his
or her efforts.
This seventh edition retains the basic organization of the sixth edition, as follows:
Part I is an overview of the research process, including developing the problem, using
the literature, preparing a research plan, and understanding ethical issues in research
and writing.
Part II introduces statistical and measurement issues in research, including statistical
descriptions, power, relationships among variables, differences among groups,
nonparametric procedures, and measurement issues in research.
Part III presents the types of, or approaches to, research, including historical,
philosophical, research synthesis, survey, descriptive, epidemiological, experimental,
12
qualitative, and mixed methods.
Part IV will help you complete the research process, which includes writing the
results and discussion, organizing the research paper, developing good figures and
tables, and presenting research in written and oral forms.
The appendix includes statistical tables.
Instructors using this text in their courses will find an instructor guide, test package,
presentation package, and image bank at
www.HumanKinetics.com/ResearchMethodsInPhysicalActivity. The image bank includes
most of the art, tables, and example elements from the text, which can be used to create
custom presentations. The instructor guide includes chapter overviews, sample course
syllabi, supplemental class activities, and student handouts. The test bank includes 302
questions.
Although the format of the book remains similar to that of the sixth edition, we have made
a number of changes that we hope improve and update the text. Various people have read
and reviewed the previous edition and provided helpful comments, including the many
students we have taught and faculty who use the book to teach research methods. We truly
do pay attention to the things you tell us. But sometimes, when we read reviews, we feel as
Day (1983, p. xi) did when he read that a reviewer described his book as both good and
original, but then went on to say that “the part that is good is not original and the part that
is original is not good.” Following is a short review of the changes in this seventh edition:
Part I: Overview of the Research Process. Each chapter includes minor revisions that
reflect updated information and more recent reports. We have made a significant
revision to chapter 2 about using library techniques by adding much more on
electronic searches. In addition, chapter 5 on ethical issues has been updated with
particular attention to procedures for the use of human and animal subjects.
Part II: Statistical and Measurement Concepts in Research. We strive in each edition to
increase the relevance of the examples and provide easy-to-understand calculations for
basic statistics. We have reduced the examples of hand calculations and formulas and
replaced them with sample output from the Statistical Package for the Social
Sciences. We have taken 2008 player performance data for outcome and skill
variables from the Professional Golfers Association website as examples for analysis in
the statistical chapters. Along with the chapter examples, the learning activities in the
instructor guide should help students grasp the fundamentals of statistical techniques.
We continue to use a unified approach to parametric and nonparametric techniques.
Part III: Types of Research. We have continued our use of expert authors to present
coherent views of sociohistorical research (a new chapter by David Wiggins and
Daniel Mason), philosophical research (R. Scott Kretchmar and Tim Elcombe), and
epidemiological research in physical activity (Barbara Ainsworth and Charles
Matthews). These three types of research are outside our expertise, and we wanted
them presented by expert scholars. Another significant revision in part III is the
13
qualitative research chapter. The many changes in the field have prompted us to
expand and update our discussion of qualitative research. Because of the increasing
simultaneous use of both quantitative and qualitative approaches within a single
study, we have expanded chapter 20 on mixed-methods research. In addition, we
have made minor revisions and updates to all the other chapters in this part.
Part IV: Writing the Research Report. The two chapters in this section remain
essentially the same, with changes and updates. Chapter 21 has a substantial new
section that includes these heads: Thesis and Dissertation Proposal, Advisor and
Dissertation Committee, The Good Scholar Must Research and Write, Scientific
Writing, and First Things Are Sometimes Best Done Last.
As we have said in each edition, we are grateful for the help of our friends, both for help
that we acknowledge in various places in the book and for help in other places where we
have inadvertently taken an idea without giving credit.
After the passage of time, one can no longer remember who originated what idea.
After the passage of even more time, it seems to me that all of the really good ideas
originated with me, a proposition which I know is indefensible (Day, 1983, p. xv).
We believe that this book provides the necessary information for both the consumer and
the producer of research. Although no amount of knowledge about the tools of research can
replace expertise in the content area, good scholars of physical activity cannot function
apart from the effective use of research tools. Researchers, teachers, clinicians, technicians,
health workers, exercise leaders, sport managers, athletic counselors, and coaches need to
understand the research process. If they do not, they are forced to accept information at
face value or on the recommendation of others. Neither is necessarily bad, but the ability to
evaluate and reach a valid conclusion based on data, method, and logic is the mark of a
professional.
Inserted into some chapters are humorous stories, anecdotes, sketches, laws, and corollaries.
These are intended to make a point and enliven the reading without distracting from the
content. Research processes are not mysterious events that graduate students should fear.
To the contrary, they are useful tools that every academic and professional should have
access to; they are, in fact, the very basis on which we make competent decisions.
Jerry R. Thomas
Jack K. Nelson
Stephen J. Silverman
14
Study Tips
Dear Student of Research Methods:
We want you to learn the material here, and most of you are learning it in a classroom
setting as well as by reading the textbook. From many years of teaching research methods,
we have arrived at the following recommendations:
The following list will help you determine your readiness to be a student of research
methods.
Score ONE point for each of the following statements that describes you:
Your library carrel is better decorated than your apartment is.
You have taken a scholarly article to a bar or coffee shop.
You rate coffee shops on the availability of outlets for your electronic devices.
You have discussed academic matters at a sporting event.
You actually have a preference between microfilm and microfiche.
You always read the reference lists in research articles.
You think that the sorority sweatshirt Greek letters are a statistical formula.
You need to explain to children why you are in the 20th grade.
You refer to stories as “Snow White et al.”
You wonder how to cite talking to yourself in APA style.
Scoring scale
Humorously yours,
15
Acknowledgments
As with any work, numerous people contributed to this book, and we want to recognize
them. Many are former students and colleagues who have said or done things that better
developed our ideas as expressed in these pages. Also, a number of faculty members who
have used previous editions have either written reviews or made suggestions that have
improved the book. Although we cannot list or even recall all these contributions, we do
know that you made them, and we thank all of you.
In particular, we thank Karen French at the University of South Carolina, Dick Magill at
Louisiana State University (emeritus), Brad Cardinal at Oregon State University, and Kathi
Thomas at the University of North Texas for allowing us to use materials that were
published jointly with them. Scott Kretchmar, Tim Elcombe, David Wiggins, Daniel
Mason, Barb Ainsworth, and Chuck Matthews made invaluable contributions with their
chapters on research methods in the areas of philosophy, history, and exercise
epidemiology, which are areas we simply could not write about effectively.
If you adopt this book for your class, we hope that you will make use of the class teaching
materials available on the Human Kinetics website at
www.HumanKinetics.com/ResearchMethodsInPhysicalActivity. Included are over 300
Microsoft PowerPoint slides, learning activities, test questions, and other course materials.
We thank Phil Martin from Iowa State University for his contributions to these materials.
Finally, we thank the staff at Human Kinetics, in particular Amy Tocco, our acquisitions
editor, Melissa Zavala, our developmental editor, and Carly S. O’Connor, our managing
editor for this edition, for their support and contributions. They have sharpened our
thinking and improved our writing.
Jerry R. Thomas
Jack K. Nelson
Stephen J. Silverman
16
Part I
Overview of the Research Process
The researches of many have thrown much darkness on the subject and if they
continue, soon we shall know nothing at all about it.
This part provides an overall perspective of the research process. The introductory chapter
defines and reviews the types of research done in physical activity and provides some
examples. We define science as systematic inquiry, and we discuss the steps in the scientific
method. This logical method answers the following four questions (Day, 1983, p. 4), which
constitute the parts of a typical thesis, dissertation, or research report:
We also present alternative approaches for doing research relative to a more philosophical
discussion of science and ways of knowing. In particular, we address qualitative research,
the use of field studies, and methods of introspection as strategies for answering research
questions instead of relying on the traditional scientific paradigm as the only approach to
research problems.
Chapter 2 suggests ways of developing a problem and using the literature to clarify the
research problem, specify hypotheses, and develop the methodology. In particular, we
emphasize the use of new electronic technology for searching, reading, analyzing,
synthesizing, organizing, and writing literature reviews.
The next two chapters in part I present the format of the research proposal with examples.
This information is typically required of the master’s or doctoral student before collecting
data for the thesis or dissertation. Chapter 3 addresses defining and delimiting the research
problem, including the introduction, statement of the problem, research hypotheses,
operational definitions, assumptions and limitations, and significance. Information is
provided for both quantitative and qualitative approaches. Chapter 4 covers methodology,
or how to do the research using either quantitative or qualitative methods. Included are the
topics of participant selection, instrumentation or apparatuses, procedures, and design and
analysis. We emphasize the value of pilot work conducted before the research and how
cause and effect may be established.
17
Chapter 5 discusses ethical issues in research and scholarship. We include information on
misconduct in science; ethical considerations in research writing, working with advisors,
and copyright; and the use of humans and animals in research.
When you have completed part I, you should better understand the research process. Then
comes the tricky part: learning all the details. We consider these details in part II (Statistical
and Measurement Concepts in Research), part III (Types of Research), and part IV
(Writing the Research Report).
18
Chapter 1
Introduction to Research in Physical
Activity
Everything that can be invented has been invented.
To each person, the word research conjures up a different picture. One might think of
searching the Internet or going to the library; another might visualize a lab filled with test
tubes, vials, and perhaps little white rats. Therefore, as we begin a text on the subject, we
must establish a common understanding of research. In this chapter, we introduce you to
the nature of research. We do this by discussing methods of problem solving and types of
research. We explain the research process and relate it to the parts of a thesis. By the time
you reach the end of chapter 1, you should understand what research really involves.
19
The Nature of Research
The object of research is to determine how things are as compared to how they might be.
To achieve this, research implies a careful and systematic means of solving problems and
involves the following five characteristics (Tuckman, 1978):
Problems to be solved come from many sources and can entail resolving controversial
issues, testing theories, and trying to improve present practice. For example, a popular topic
of concern is obesity and methods of losing weight. Suppose we want to investigate this
issue by comparing the effectiveness of two exercise programs in reducing fat. Of course, we
know that caloric expenditure results in the loss of fat, so we will try to find out which
program does this better under specified conditions. (Note: Our approach here is to give a
simple, concise overview of a research study. We do not intend it to be a model of
originality or sophistication.)
This study is an example of applied research. Rather than try to measure the calories
expended and so on, we approach it strictly from a programmatic standpoint. Say that we
are operating a health club and offer aerobic dance and jogging classes for people who want
to lose weight. Our research question is this: Which program is more effective in reducing
fat?
applied research—A type of research that has direct value to practitioners but in which the
researcher has limited control over the research setting.
Suppose we have a pool of participants to draw from and can randomly assign two-thirds of
them to the two exercise programs and one-third to a control group. We have their scout’s
honor that no one is on a drastic diet or is engaging in any other strenuous activities while
the study is in progress. Both the aerobic dance class and the jogging class are one hour
long and are held five times a week for 10 weeks. The same enthusiastic and immensely
qualified instructor teaches both classes.
20
Our measure of fatness is the sum of skinfold measurements taken at eight body sites. Of
course, we could use other measures, such as the percentage of fat estimated from
hydrostatic weighing (or total body water or some other estimate of fatness). In any case,
we can defend our measures as valid and reliable indicators of fatness, and skinfolds are
functional field measures. We measure all the participants, including those in the control
group, at the beginning and the end of the 10-week period. During the study we try to
ensure that the two programs are similar in procedural aspects, such as motivational
techniques and the aesthetics of the surroundings. In other words, we do not favor one
group by cheering them on and not encouraging the other, nor do we have one group
exercise in an air-conditioned, cheerful, and healthful facility while the other has to sweat it
out in a dingy room or a parking lot. We try to make the programs as similar as possible in
every respect except the experimental treatments. The control group does not engage in any
regular exercise.
hydrostatic weighing—A technique for measuring body composition in which body
density is computed by the ratio of a person’s weight in air and the loss of weight
underwater.
After we have measured all the participants on our criterion of fatness at the end of the 10-
week program, we are ready to analyze our data. We want to see how much change in
skinfold thickness has occurred and whether differences have occurred between the two
types of exercise. Because we are dealing with samples of people (from a whole universe of
similar people), we need to use some type of statistics to establish how confident we can be
in our results. In other words, we need to determine the significance of our results. Suppose
the mean (average) scores for the groups are as follows:
These values (which we made up) represent the average change in the combined skinfold
thicknesses of the eight body sites. The two experimental groups lost fat, but the control
group actually showed increased skinfold thicknesses over the 10-week period.
We decide to use the statistical technique of analysis of variance with repeated measures.
We find a significant F ratio, indicating that significant differences exist among the three
groups. Using a follow-up test procedure, we discover that both exercise groups are
significantly different from the control group. But we find no significant difference between
the aerobic dance and the jogging groups. (Many of you may not have the foggiest idea
what we are talking about with the statistical terms F ratio and significance, but do not
worry about it. All that is explained later. This book is directly concerned with those kinds
of things.)
Our conclusion from this study is that both aerobic dance and jogging are effective
21
(apparently equally so) in bringing about a loss in fatness of overweight people (such as the
ones in our study) over 10 weeks. Although these results are reasonable, remember that this
scenario is only an example. We could also pretend that this study was published in a
prestigious journal and that we won the Nobel Prize.
Research Continuum
Research in our field can be placed on a continuum that has applied research at one
extreme and basic research at the opposite extreme. The research extremes are generally
associated with certain characteristics. Applied research tends to address immediate
problems, use so-called real-world settings, use human participants, and involve limited
control over the research setting, but it gives results that are of direct value to practitioners.
At the other extreme, basic research usually deals with theoretical problems. It uses the
laboratory as the setting, frequently uses animals as subjects, carefully controls conditions,
and produces results that have limited direct application. Christina (1989) suggested that
basic and applied forms of research are useful in informing each other as to future research
directions. Table 1.1 demonstrates how research problems in motor learning might vary
along a continuum from basic to applied depending on their goals and approaches.
basic research—A type of research that may have limited direct application but in which
the researcher has careful control of the conditions.
22
To some extent, the strengths of applied research are the weaknesses of basic research, and
vice versa. Considerable controversy exists in the literature on social science (e.g., Creswell,
2009; Jewczyn, 2013) and physical activity (e.g., Christina, 1989) about whether research
should be more basic or more applied. This issue, labeled ecological validity, deals with
two concerns: Is the research setting perceived by the research participant in the way
intended by the experimenter? Does the setting have enough of the real-world
characteristics to allow generalizing to reality?
ecological validity—The extent to which research emulates the real world.
Of course, most research is neither purely applied nor purely basic; rather, it incorporates
some degree of both. We believe that systematic efforts are needed in the study of physical
activity to produce research that moves back and forth across Christina’s (1989) levels of
research (table 1.1). Excellent summaries of this type of research and the accumulated
knowledge are provided in three edited volumes representing exercise physiology, sport
psychology, and motor behavior: Physical Activity, Fitness, and Health (Bouchard, Shepard,
& Stephens, 1994), Handbook of Sport Psychology (Singer, Hausenblas, & Janelle, 2001),
and Cognitive Issues in Motor Expertise (Starkes & Allard, 1993). An expert prepared each
chapter in these books to summarize theories as well as to present basic and applied research
about areas related to exercise physiology, sport psychology, and motor expertise. The
novice researcher would do well to read several of these chapters as examples of how
knowledge is developed and accumulated in the study of physical activity. We need more
efforts to produce a related body of knowledge in the study of physical activity. Although
the research base has grown tremendously in our field over the past 40 years, much remains
to be done.
23
(as well as in the fields of psychology, sociology, education, and physiology). Quality
research efforts always involve some or all of the following actions:
All kidding aside, the need for research in any profession cannot be denied. After all, one of
the primary distinctions between a discipline or profession and a trade is that the trade
deals only with how to do something, whereas the discipline or profession concerns itself
not only with how but also with why something should be done in a certain manner (and
why it should even be done at all). But although most people in a discipline or profession
recognize the need for research, most do not read research results. This situation is not
unique to our field. It has been reported that only 1% of chemists read research
publications, that fewer than 7% of psychologists read psychological research journals, and
so on. The big question is why. We guess that most professionals who do not read research
believe that doing so is not necessary. They believe that research is not practical enough or
does not directly pertain to their work. Another reason given by practitioners for not
reading research publications is that they cannot understand them. The language is too
technical, and the terminology is unfamiliar and confusing. This complaint is valid, but we
could argue that if the professional preparation programs were more scientifically oriented,
24
the problem would diminish. Nevertheless, the research literature is extremely difficult for
the nonresearcher to understand.
Reading Research
Someone once said (facetiously) that scientific papers are meant not to be read but to be
published. Unfortunately, we find considerable truth in this observation. We writers are
often guilty of trying to use language to dazzle the reader and perhaps to give the
impression that our subject matter is more esoteric than it really is. We tend to write for the
benefit of a rather small number of readers—that is, other researchers in our field.
We have the problem of jargon, of course. In any field, whether it is physics, football, or
cake baking, jargon confounds the outsider. The use of jargon serves as a kind of
shorthand. It provides meaning to the people within the field because everyone uses those
words in the same context. Research literature is famous for using a three-dollar word when
a nickel word would do. As Day and Gastel (2006, p. 200) asked, what self-respecting
writer would use a three-letter word such as now when he or she could use the elegant
expression at this point in time? Researchers never do anything, they perform it; they never
start, they initiate; and they terminate instead of end. Day and Gastel further remarked that
an occasional author slips and uses the word drug, but most salivate like Pavlov’s dogs in
anticipation of using chemotherapeutic agent.
The need to bridge the gap between the researcher and the practitioner has been recognized
for years. For example, the Journal of Physical Education, Recreation and Dance has a feature
called “Research Works,” which disseminates applied research information to teachers,
coaches, and fitness and recreation leaders. The website for the American Kinesiology
Association (www.americankinesiology.org) regularly has a section on applied research. Yet
despite these and other attempts to bridge the gap between researchers and practitioners,
the gap is still imposing.
It goes without saying that if you are not knowledgeable about the subject matter, you
cannot read the research literature. Conversely, if you know the subject matter, you can
probably wade through the researcher’s jargon more effectively. For example, if you know
baseball and the researcher is recommending that by shortening the radius, the hitter can
increase the angular velocity, you can figure out that the researcher means to choke up on
the bat.
One of the big stumbling blocks is the statistical analysis part of research reports. Even the
most ardent seeker of knowledge can be turned off by such descriptions as this: “The
tetrachoric correlations among the test variables were subjected to a centroid factor analysis,
and orthogonal rotations of the primary axes were accomplished by Zimmerman’s graphical
method until simple structure and positive manifold were closely approximated.” Please
25
note that we are not criticizing the authors for such descriptions, because reviewers and
editors usually require them. We are just acknowledging that statistical analysis is
frightening to someone who is trying to read a research article and does not know a factor
analysis from a volleyball. The widespread use of computers and “computerese” probably
compounds the mystery associated with statistics. Many people believe anything that comes
out of a computer. Others are more old-fashioned and check the computer’s accuracy with
their calculators. A classic, yet fictional, case of a computer mistake occurred in a high
school in which the computer printed the students’ locker numbers in the column where
their IQs were supposed to go. It is classic because no one noticed the error at the time, but
at the end of the year the students with the highest locker numbers got the best grades.
Despite all the hurdles that loom in the practitioner’s path when reading research, we
contend that you can read and profit from (usually not materially, but then consider the
data on the new Speedo swimsuit and its influence on the 2008 swimming results in
Beijing) the research literature even if you are not well grounded in research techniques and
statistical analysis. We would like to contend that after you read this book, you will be able
to read any journal in any field, but the publisher would not let us say that. We offer the
following suggestions on reading the research literature:
Become familiar with a few publications that contain pertinent research in your field.
You might get some help on choosing the publications from a professor or librarian.
Read only studies that are of interest to you. This point may sound too trite to
mention, but some people feel obligated to wade through every article.
Read as a practitioner would. Do not look for eternal truths. Look for ideas and
indications. No study is proof of anything. Only when it has been verified repeatedly
does it constitute knowledge.
Read the abstract first. This saves time by helping you determine whether you wish to
read the whole thing. If you are still interested, then you can read the study to gain a
better understanding of the methodology and the interpretations, but do not get
bogged down with details.
Do not be too concerned about statistical significance. Understanding the concept of
significance certainly helps, but a little common sense serves you about as well as
knowing the difference between the .02 and the .01 levels, or a one-tailed test versus
a two-tailed test. Think in terms of meaningfulness. For example, if two methods of
teaching bowling result in an average difference of 0.5 pins, what does it matter
whether the difference is significant? On the other hand, if a big difference is present
but not significant, further investigation is warranted, especially if the study involved
a small number of participants. Knowing the concepts of the types of statistical
analysis is certainly helpful, but it is not crucial to being able to read a study. Just skip
that part.
26
Be critical but objective. You can usually assume that a national research journal
selects studies for publication by the jury method. Two or three qualified people read
and judge the relevance of the problem, the validity and reliability of the procedures,
the efficacy of the experimental design, and the appropriateness of the statistical
analysis. Certainly, some studies are published that should not be. Yet if you are not
an expert in research, you do not need to be suspicious about the scientific worth of a
study that appears in a recognized journal. If it is too far removed from any practical
application to your situation, do not read it.
You will find that the more you read, the more you understand, simply because you
become more familiar with the language and the methodology, like the man who was
thrilled to learn he had been speaking prose all his life.
The next section of the study dealt with methods. It noted the length of the season, the
number of games, the number of practices and their length, and the breakdown of time
devoted to drills, scrimmages, and individual practice. The participants were 12 girls on the
high school basketball team and 14 girls in the nonparticipant group who were in physical
education classes and had academic and activity schedules similar to those of the
participants. All participants in the study were tested at the beginning and end of the season
on maximal oxygen consumption and various other physiological measurements dealing
with ventilation, heart rate, and blood pressure. Sonjia remembered those things from her
exercise physiology course a number of years earlier and was willing to accept these as good
indicators of cardiorespiratory fitness.
The results were then presented in tables. Sonjia did not understand these things but was
willing to trust the author as to their appropriateness. The author discovered no significant
27
increases in any of the cardiorespiratory measures from the preseason test to the postseason
test for either group. This finding jolted Sonjia to the quick! Surely, a strenuous sport such
as basketball should produce improvements in fitness. Something must be wrong here, she
thought. She further read (with small consolation) that the basketball players had higher
values of maximal oxygen consumption than the nonparticipants did at both the beginning
and the end of the season. Sonjia then read the discussion, which mentioned things such as
that the values were higher than similar values in other studies. (So what? Sonjia thought.)
She read with more interest the observations by the author that boys’ basketball programs
were more strenuous in terms of length and number of workouts. Sonjia began to think
about this. The author admitted that the number of participants was small and that some
changes might not have been detected, and the author offered other speculations. The
author concluded, however, that the training program used in this study was not strenuous
enough to induce significant improvement in cardiorespiratory fitness.
discussion—The chapter or section of a research report that explains what the results
mean.
Sonjia was sophisticated enough to realize the limitations of the single study. Nevertheless,
the practice schedule and general practice routines used in the study were similar to her
own. She noticed in the references for the article three studies from a journal called
Medicine & Science in Sports & Exercise. She had never read that journal, but she decided to
drive over to the university the next weekend to look up the publication in the library.
When she located the journal, the latest issue happened to have an article on the
conditioning effects of swimming on college women. Although the study dealt with a
different sport and a different age group, she reasoned that the review of literature might
prove fruitful. She was right. It cited a recent study on aerobic capacity, heart rate, and
energy cost during a season of girls’ basketball. Sonjia quickly located the study and then
read with the excitement that comes from the personal discovery of ideas. She also was
pleasantly surprised to find that it was easier reading than the first study because she was
now more familiar with the terminology and the general organization of the article.
This study also reported no improvement in aerobic capacity during the season. The study
involved monitoring heart rates during games by telemetry, and researchers infrequently
observed heart rates of over 170 beats per minute (bpm). They concluded that the practice
sessions were apparently too moderate in intensity and that the training should be
structured to meet both the skill and the fitness demands of the sport.
Sonjia returned to her school determined to take a more scientific approach to her
basketball program. To start, she had one of her managers chart the number of minutes
players were actually engaged in movement in the practice sessions. Sonjia also had the
players take their pulses at various intervals during the sessions. She was surprised to find
that the heart rates rarely surpassed 130 bpm. As an outgrowth of her recent literature
search, she remembered that an intensity threshold was necessary to bring about
28
improvement in cardiorespiratory fitness. She knew that for this age group a heart rate of
about 160 bpm was needed to provide a significant training effect. Consequently, she
initiated some changes in her practice sessions (including more conditioning drills) and
made the scrimmages more intensive and gamelike. To end this saga of Sonjia, you will be
happy to know that Coach Roundball’s team went on to win all its games, the district and
the state championships, and the world games.
Having said that about science, discovery can be rewarding, whether that discovery is
research that applies to and can improve your situation or new knowledge obtained while
researching your thesis or dissertation. Research should be viewed more as a method of
problem solving than as some dark and mysterious realm inhabited by impractical people
who speak and write in baffling terms. We believe that practitioners can read research
literature, and we are dedicated in this text to facilitating the process of becoming a research
consumer.
29
Unscientific Versus Scientific Methods of Problem
Solving
Although there are many definitions of research, nearly all characterize research activity as
some sort of structured problem solving. The word structured refers to the fact that a
number of research techniques can be used as long as they are considered acceptable by
scholars in the field. Thus, research is concerned with problem solving, which then may
lead to new knowledge.
The problem-solving process involves several steps whereby the problem is developed,
defined, and delimited; hypotheses are formulated; data are gathered and analyzed; and the
results are interpreted with regard to the acceptance or rejection of the hypotheses. These
steps are often referred to as the scientific method of problem solving. The steps also
constitute the chapters, or sections, of the research paper, thesis, or dissertation.
Consequently, we devote much of this text to the specific ways these steps are
accomplished.
scientific method of problem solving—A method of solving problems that uses the
following steps: defining and delimiting the problem, forming a hypothesis, gathering
data, analyzing data, and interpreting the results.
Tenacity
People sometimes cling to certain beliefs despite a lack of supporting evidence. Our
superstitions are good examples of the method called tenacity. Coaches and athletes are
notoriously superstitious. A coach may wear a particular sport coat, hat, tie, or pair of shoes
because the team won the last time he wore it. Athletes frequently have a set pattern that
they consider lucky for dressing, warming up, or entering the stadium. Although they
acknowledge no logical relationship between the game’s outcome and their particular
routine, they are afraid to break the pattern.
tenacity—An unscientific method of problem solving in which people cling to certain
beliefs regardless of a lack of supporting evidence.
30
For example, take the man who believed that black cats bring bad luck. One night when he
was returning to his ranch, a black cat started to cross the road. The man swerved off onto
the prairie to keep the cat from crossing in front of him and hit a hard bump that caused
the headlights to turn off. Unable to see the black cat in the dark night, he sped frantically
over rocks, mounds, and holes until he came to a sudden stop in a ravine, wrecking his car
and sustaining moderate injuries. Of course, this episode just confirmed his staunch belief
that black cats do indeed bring bad luck.
Obviously, tenacity has no place in science. It is the least reliable source of knowledge.
Intuition
Authority
Reference to some authority has long been used as a source of knowledge. Although this
approach is not necessarily invalid, it does depend on the authority and the rigidity of
adherence. But the appeal to authority has been carried to absurd lengths. Even personal
observation and experience have been deemed unacceptable when they dispute authority.
Supposedly, people refused to look through Galileo’s telescope when he disputed Ptolemy’s
explanation of the world and the heavens. Galileo was later jailed and forced to recant his
beliefs. Bruno also rejected Ptolemy’s theory and was burned at the stake. (Scholars read
and believed Ptolemy’s book on astrology and astronomy for 1,200 years after his death!) In
1543 Vesalius wrote a book on anatomy, much of which is still considered correct today.
But because his work clashed with Galen’s theories, he met with such ridicule that he gave
up his study of anatomy.
Perhaps the most crucial aspect of the appeal to authority as a means of obtaining
knowledge is the right to question and to accept or reject the information. Furthermore,
the authority’s qualifications and the methods by which the authority acquired the
knowledge also determine the validity of this source of information.
Rationalistic Method
31
In the rationalistic method, we derive knowledge through reasoning. A good example is the
following classic syllogism:
Although you probably would not argue with this reasoning, the key to this method is the
truth of the premises and their relationship to each other. For example,
In this case, however, Tom is very short. The conclusion is trustworthy only if it is derived
from premises (assumptions) that are true. Also, the premises may not in fact be premises
but rather descriptions of events or statements of fact. The statements are not connected in
a cause-and-effect manner. Consider the following example:
Of course, in the first statement the factor common to both mathematics achievement and
shoe size is age. Older children tend to be bigger and thus have bigger feet than younger
children. Older children also have higher achievement scores in mathematics, but there is
no cause-and-effect relationship. You must always be aware of this when dealing with
correlation. Reasoning is fundamental in the scientific method of problem solving but
cannot be used by itself to arrive at knowledge.
Empirical Method
The word empirical denotes experience and the gathering of data. Certainly, data gathering
is part of the scientific method of solving problems. But relying too much on your own
experience (or data) has drawbacks. First, your own experience is limited. Furthermore,
your retention depends substantially on how the events agree with your experience and
beliefs, on whether things “make sense,” and on your state of motivation to remember.
Nevertheless, the use of data (and the empirical method) is high on the continuum of
methods of obtaining knowledge as long as you are aware of the limitations of relying too
heavily on this method.
32
empirical—A description of data or a study that is based on objective observations.
This step may sound contradictory, because how could the development of the problem be
part of solving it? Actually, the discussion here is not about finding a problem to study
(ways of locating a problem are discussed in chapter 2); the assumption is that the
researcher has already selected a topic. But to design and execute a sound investigation, the
researcher must be specific about what is to be studied and to what extent it will be studied.
Many ramifications constitute this step, an important one being the identification of the
independent and dependent variables. The independent variable is what the researcher is
manipulating. If, for example, two methods of teaching a motor skill are being compared,
then the teaching method is the independent variable; this item is sometimes called the
experimental, or treatment, variable.
independent variable—The part of the experiment that the researcher is manipulating;
also called the experimental variable or treatment variable.
The dependent variable is the effect of the independent variable. In the comparison of
teaching methods, the measure of skill is the dependent variable. If you think of an
experiment as a cause-and-effect proposition, the cause is the independent variable and the
effect is the dependent variable. The latter is sometimes referred to as the yield. Thus, the
researcher must define exactly what will be studied and what will be the measured effect.
When this question is resolved, the experimental design can be determined.
dependent variable—The effect of the independent variable; also called the yield.
The hypothesis is the expected result. A person setting out to conduct a study generally has
an idea as to what the outcome will be. This anticipated solution to the problem may be
based on some theoretical construct, on the results of previous studies, or perhaps on the
33
experimenter’s experience and observations. The last source is probably least likely or
defensible because of the weaknesses of the unscientific methods of acquiring knowledge
discussed previously. Regardless, the research should have some experimental hypothesis
about each subproblem in the study.
hypothesis—The anticipated outcome of a study or experiment.
We enjoyed Calvin and Hobbes, the cartoon strip by Bill Watterson. In a clever strip,
Calvin is talking to his friend Susie in the lunchroom:
Calvin: Curiosity is the essence of the scientific mind. For example, you know how
milk comes out your nose if you laugh while drinking? Well, I’m going to see what
happens when I inhale milk into my nose and laugh!
Susie (as she leaves): Idiocy is the essence of the male mind.
Calvin: I’m guessing it will shoot out my ears. Don’t you want to see?
Calvin has developed a testable hypothesis: If I inhale milk into my nose and laugh, it will
shoot out of my ears. Susie has an untestable hypothesis (at least in our view): “Idiocy is the
essence of the male mind.”
One of the essential features about the hypothesis is that it be testable. The study must be
designed in such a way that the hypothesis can be either supported or refuted. Obviously,
then, the hypothesis cannot be a type of value judgment or an abstract phenomenon that
cannot be observed.
For example, you might hypothesize that success in athletics depends solely on fate. In
other words, if a team wins, it is because it was meant to be; similarly, if a team loses, a
victory was just not meant to be. Refuting this hypothesis is impossible because no evidence
could be obtained to test it.
Of course, before step 2 can be accomplished, the researcher must decide on the proper
methods of acquiring the necessary data to be used in testing the research hypothesis. The
reliability of the measuring instruments, the controls that are employed, and the overall
objectivity and precision of the data-gathering process are crucial to solving the problem.
In terms of difficulty, gathering data may be the easiest step because in many cases it is
routine. Planning the method, however, is one of the most difficult steps. Good methods
attempt to maximize both the internal validity and the external validity of the study.
34
internal validity—The extent to which the results of a study can be attributed to the
treatments used in the study.
external validity—The generalizability of the results of a study.
Internal validity and external validity relate to the research design and controls that are
used. Internal validity refers to the extent to which the results can be attributed to the
treatments used in the study. In other words, the researcher must try to control all other
variables that could influence the results. For example, Jim Nasium wants to assess the
effectiveness of his exercise program in developing physical fitness in young boys. He tests
his participants at the beginning and then at the end of a 9-month training program and
concludes that the program brought about significant improvement in fitness. What is
wrong with Jim’s conclusion? His study contains several flaws. The first is that Jim did not
consider maturity. Nine months of maturation produced significant changes in size and in
accompanying strength and endurance. Also, what else were the participants doing during
this time? How do we know that other activities were not responsible, or partly so, for the
changes in their fitness levels? Chapter 18 deals with these threats to internal validity.
External validity pertains to the generalizability of the results. To what extent can the
results apply to the real world? A paradox often occurs for research in the behavioral
sciences because of the controls required for internal validity. In motor-learning studies, for
example, the task is often something novel so that it provides a control for experience.
Furthermore, being able to measure the performance objectively and reliably is desirable.
Consequently, the learning task is frequently a maze, a rotary pursuit meter, or a linear
position task, all of which may meet the demands for control with regard to internal
validity. But then you face the question of external validity: How does performance in a
laboratory setting with a novel, irrelevant task apply to learning gymnastics or basketball?
These questions are important and sometimes vexing, but they are not insurmountable.
(They are discussed later.)
We will never run out of math professors because they always multiply.
When the body is fully immersed in water, the telephone rings.
If there are only two people in a locker room, they will have adjacent lockers.
The ocean would be much deeper without sponges.
35
The novice researcher finds this step to be the most formidable for several reasons. First,
this step usually involves some statistical analysis, and the novice researcher (particularly the
master’s student) often has a limited background in and a fear of statistics. Second, analysis
and interpretation require considerable knowledge, experience, and insight, which the
novice may lack.
That analyzing and interpreting results is the most challenging step goes without question.
It is here that the researcher must provide evidence for the support or rejection of the
research hypothesis. In doing this, the researcher also compares the results with those of
others (the related literature) and perhaps attempts to relate and integrate the results into
some theoretical model. Inductive reasoning is employed in this step (whereas deductive
reasoning is primarily used in the statement of the problem). The researcher attempts to
synthesize the data from his or her study along with the results of other studies to
contribute to the development or substantiation of a theory.
36
Alternative Models of Research
In the preceding section we summarized the basic steps in the scientific method of problem
solving. Science is a way of knowing, often defined as structured inquiry. One basic goal of
science is to explain things or to generalize and build a theory. When a scientist develops a
useful model to explain behavior, scholars often test predictions from this model using the
steps of the scientific method. The model and the approaches used to test the model are
called a paradigm.
Normal Science
For centuries, the scientific approaches used in studying problems in both the natural and
the social sciences have been what Thomas Kuhn (1970), a noted science historian, termed
normal science. This manner of study is characterized by the elements we listed at the
beginning of this chapter (i.e., systematic, logical, empirical, reductive, and replicable). Its
basic doctrine is objectivity. Normal science is grounded in the natural sciences, which have
long adhered to the idea of the orderliness and reality of matter—that is, that nature’s laws
are absolute and discoverable by objective, systematic observations and investigations that
are not influenced by (in other words, independent of) humans. The experiments are
theory driven and have testable hypotheses.
normal science—An objective manner of study grounded in the natural sciences that is
systematic, logical, empirical, reductive, and replicable.
Normal science received a terrific jolt with Einstein’s theory of relativity and the quantum
theory, which indicated that nature’s laws could be influenced by humans (that is, that
reality depends to a great extent on how one perceives it). Moreover, some things, such as
the decay of a radioactive nucleus, happen for no reason at all. The fundamental laws that
had been believed to be absolute were now considered statistical rather than deterministic.
Phenomena could be predicted statistically but not explained deterministically (Jones,
1988).
Relatively recently (since about 1960), serious challenges have arisen regarding normal
science’s concept of objectivity (i.e., that the researcher can be detached from the
instruments and conduct of the experiment). Two of the most powerful challengers to the
idea of objective knowledge were Thomas Kuhn (1970) and Michael Polanyi (1958). They
contended that objectivity is a myth. From the first inception of the idea for the hypothesis
through the selection of apparatus to the analysis of the results, the observer is involved.
37
The conduct of the experiment and the results can be considered expressions of the
researcher’s point of view. Polanyi was especially opposed to the adoption of normal science
for the study of human behavior.
Kuhn (1970) maintained that normal science does not really evolve in systematic steps the
way that scientific writers describe it. Kuhn discussed the paradigm crisis phenomenon, in
which researchers who have been following a particular paradigm begin to find
discrepancies in it. The findings no longer agree with the predictions, and a new paradigm
is advanced. Interestingly, the old paradigm does not die completely but only develops
varicose veins and fades away. Many researchers with a great deal of time and effort
invested in the old paradigm are reluctant to change, so it is usually a new group of
researchers who propose the new paradigm. Thus, normal science progresses by revolution,
with a new group of scientists breaking away and replacing the old. Kuhn and Polanyi
concurred that the doctrine of objectivity is simply not a reality. Nevertheless, normal
science has been and will continue to be successful in the natural sciences and in certain
aspects of the study of humans. But Martens (1987) contended that it has failed miserably
in the study of human behavior, especially in the more complex functions.
paradigm crisis phenomenon—The development of discrepancies in a paradigm leading to
proposals of a new paradigm that better explains the data.
As a sport psychologist, Martens asserted that laboratory experiments have limited use in
answering questions about complex human behavior in sport. He considered his role as a
practicing sport psychologist to have been far more productive in gaining knowledge about
athletes and coaches and the solutions to their problems. Other workers in the so-called
helping professions have made similar observations about both the limitations of normal
science and the importance of alternative sources of knowledge in forming and shaping
professional beliefs. Schein (1987), a noted scholar of social psychology, related an
interesting (some might call it shocking) revelation concerning the relative influence of
published research results versus practical experience. At a conference, he and a number of
his colleagues were discussing what they relied on most for their classroom teaching. These
professors seemed to agree that the data they really believed in and used in the classroom
came from personal experience and information learned in the field. Schein was making the
point that different categories of knowledge can be obtained by different methods. In
effect, some people are more influenced by sociological and anthropological research
models than by the normal science approach.
For some time, many scholars in education, psychology, sociology, anthropology, sport
psychology, physical education, and other disciplines have proposed methods of studying
human behavior other than those of conventional normal science. Anthropologists,
sociologists, and clinical psychologists have used in-depth observation, description, and
analysis of human behavior for nearly three-quarters of a century. For over 40 years,
researchers in education have used participant and nonparticipant observation to obtain
38
comprehensive, firsthand accounts of teacher and student behaviors as they occur in real-
world settings. More recently, physical educators, sport psychologists, and exercise
specialists have been engaged in this type of field research. A number of names given to this
general form of research are ethnographic, qualitative, grounded, naturalistic, and
participant observational research. Regardless of the names, the commitments, and the
beliefs of the researchers, this type of research was not well received at first by the adherents
of normal science and the scientific method. In fact, this form of research (we include all its
forms under the name qualitative research) has often been labeled by normal scientists as
superficial, lacking in rigor, and just plain unscientific. As qualitative research methods have
evolved, so has the thinking of many of these people. As you will see in chapters 19 and 20,
many of the research tenets listed by Kuhn (1970) are found in contemporary qualitative
research.
qualitative research—A research method that often involves intensive, long-term
observation in a natural setting; precise and detailed recording of what happens in the
setting; and the interpretation and analysis of the data using description, narratives,
quotes, charts, and tables. Also called ethnographic, naturalistic, interpretive, grounded,
phenomenological, subjective, and participant observational research.
The debates between qualitative and normal (often classified as quantitative) research have
been heated and prolonged. The qualitative proponents have gained confidence and
momentum in recent years, and this point of view is now recognized as a viable method of
addressing problems in the behavioral sciences. Credibility is established by systematically
categorizing and analyzing causal and consequential factors. The naturalistic setting of
qualitative research both facilitates analysis and precludes precise control of so-called
extraneous factors, as does much other research occurring in field settings. The holistic
interrelationship among observations and the complexity and dynamic processes of human
interaction make it impossible to limit the study of human behavior to the sterile,
reductionistic approach of normal science. Reductionism, a characteristic of normal
science, assumes that complex behavior can be reduced, analyzed, and explained as parts
that can then be put back together as a whole and understood. Critics of the conventional
approach to research believe that the central issue is the unjustified belief that normal
science is the only source of true knowledge.
reductionism—A characteristic of normal science that assumes that complex behavior can
be reduced, analyzed, and explained as parts that can then be put back together to
39
understand the whole.
The challenges to normal science involve many implications. For example, when we study
simple movements, such as linear positioning in a laboratory to reflect cognitive processing
of information, do we learn anything about movements in real-world settings such as the
performance of sport skills? When we evaluate EMG activity in specific muscle groups
during a simple movement, does the result really tell us anything about the way the nervous
system controls movements in natural settings such as athletics? Can we study the
association of psychological processes related to movement in laboratory settings and expect
the results to apply in sport and exercise situations? When we conduct these types of
experiments, are we studying nature’s phenomena or laboratory phenomena?
Do not misinterpret the intent of these questions. They do not mean that nothing
important can be discovered about physical activity from laboratory research. What they
suggest is that these findings do not necessarily accurately model the way humans plan,
control, and execute movements in natural settings associated with exercise and sport.
Kuhn’s (1970) explanations of how science advances and the limitations of applying
normal science to natural settings demonstrate that scientists need to consider the various
ways of knowing and that the strict application of the normal scientific method of problem
solving may sometimes hinder rather than advance science. If the reductionistic approach of
the scientific method has not well served the natural scientists who developed it, then
certainly researchers in human behavior need to assess the relative strengths and weaknesses
of conventional and alternative research paradigms for their particular research questions.
40
Figure 1.1 The degrees of knowledge theory with examples of methods varying in degree of
reliability.
Reprinted, by permission, from R. Martens, 1987, “Science, knowledge, and sport psychology,” The Sport Psychologist
1(1): 46.
What we hope you gain from this section is that science is disciplined inquiry, not a set of
specific procedures. Although advocates of alternative methods of research are often
persuasive, we do not want you to conclude that the study of physical activity should
abandon the traditional methods of normal science. We have learned much from these
techniques and will continue to do so. Furthermore, we certainly do not want you to toss
away this book as being pointless. We have not even begun to tell you all the fascinating
things that we have learned over the years (it is hard to tell whether some of these things
should be classified as normal or abnormal science). In addition, we have many funny
stories yet to tell (abnormal humor). Aside from these compelling reasons for continuing
with the book, we want you to realize and appreciate that so-called normal science is not
the solution to all questions raised in our field. Furthermore, none of the alternative
methods of research denounces the scientific method of problem solving.
The bottom line is that different problems require different solutions. As we said before,
science is disciplined inquiry, not a set of specific procedures. We need to embrace all
systematic forms of inquiry. Rather than argue about the differences, we should capitalize
on the strengths of all scholarly methods to provide useful knowledge about human
movement. The nature of the research questions and setting should drive the selection of
approaches to acquiring knowledge. In fact, just as Christina (1989) suggested that
41
researchers might move among levels of research (basic to applied), so researchers might
move among paradigms (quantitative to qualitative to mixed methods) to acquire
knowledge. In addressing this issue, we have expanded the qualitative and mixed-method
research sections in chapters 19 and 20, which focus on using varying types of research
approaches. Of course, we do not want to be perceived like Danae in the comic strip Non
Sequitur (by Wiley). Danae (a young girl) says to her horse that she wants to grow up to be
a preconceptual scientist. Her horse asks, “What is that?” to which Danae responds that it is
“the new science of reaching a conclusion before doing any research and then simply
dismissing anything contrary to your preconceived notions.”
42
Types of Research
Research is a structured way of solving problems. Different kinds of problems attend the
study of physical activity; thus, different types of research are used to solve these problems.
This text concentrates on five types of research: analytical, descriptive, experimental,
qualitative, and mixed methods. A brief description of each follows.
Analytical Research
As the name implies, analytical research involves in-depth study and the evaluation of
available information in an attempt to explain complex phenomena. The types of analytical
research are historical, philosophical, reviews, and research synthesis.
analytical research—A type of research that involves in-depth study and the evaluation of
available information in an attempt to explain complex phenomena; can be categorized
in the following way: historical, philosophical, reviews, and research synthesis.
43
My research methods teacher will love this idea.
© Creativa/fotolia
Historical Research
Obviously, historical research deals with events that have already occurred. Historical
research focuses on events, organizations, institutions, and people. In some studies, the
researcher is interested mostly in preserving the record of events and accomplishments. In
other investigations, the researcher attempts to discover facts that will provide more
meaning and understanding of past events to explain the present state of affairs. Some
historians have even attempted to use information from the past to predict the future. The
research procedures associated with historical studies are addressed in considerable detail in
chapter 12.
Philosophical Research
Although some authors emphasize the differences between science and philosophy, the
philosophical method of research follows essentially the same steps as other methods of
scientific problem solving. The philosophical approach uses scientific facts as the basis for
formulating and testing research hypotheses.
An example of such philosophical research was Morland’s 1958 study in which he analyzed
44
the educational views held by leaders in American physical education and categorized them
into educational philosophies of reconstructionism, progressivism, essentialism, and
perennialism.
Having an opinion is not the same as having a philosophy. In philosophical research, beliefs
must be subjected to rigorous criticism in light of the fundamental assumptions. Academic
preparation in philosophy and a solid background in the fields from which the facts are
derived are necessary. Other examples and a more detailed explanation of philosophical
research are given in chapter 13.
Reviews
A review is a critical evaluation of recent research on a particular topic. The author must be
extremely knowledgeable about the available literature as well as the research topic and
procedures. A review involves an analysis, evaluation, and integration of the published
literature, often leading to important conclusions concerning the research findings up to
that time (for good examples of reviews, see Blair, 1993, and Silverman and Subramaniam,
1999).
review—A critical evaluation of research on a particular topic.
Certain publications consist entirely of reviews, such as Psychological Review, Annual Review
of Physiology, Review of Educational Research, and Kinesiology Review. A number of journals
publish reviews periodically, and some occasionally devote entire issues to reviews. For
example, the 75th anniversary issue of Research Quarterly for Exercise and Sport (Silverman,
2005) contains some excellent reviews on various topics.
Research Synthesis
Reviews of literature are difficult to write because they require the synthesis of a large
number of studies to determine common underlying findings, agreements, or
disagreements. To some extent this is like trying to make sense of data collected on a large
number of participants by simply looking at the data. Glass (1977) and Glass, McGaw, and
Smith (1981) proposed a quantitative means of analyzing the findings from numerous
studies; this method is called meta-analysis. Findings between studies are compared by
changing results within studies to a common metric called effect size. Over the years, many
meta-analyses have been reported in the physical activity literature (e.g., Payne & Morrow,
1993; Rawdon, Sharp, Shelley & Thomas, 2012; Schieffer & Thomas, 2012; Thomas &
French, 1985). This technique is discussed in more detail in chapter 14.
Descriptive Research
45
Descriptive research is concerned with status. The most prevalent descriptive research
technique is the survey, most notably the questionnaire. Other forms of surveys include the
interview (personal and by telephone) and the normative survey. Chapter 15 provides
detailed coverage of these techniques. The following sections briefly describe three types of
survey research techniques.
descriptive research—A type of research that attempts to describe the status of the study’s
focus. Common techniques are questionnaires, interviews, normative surveys, case
studies, job analyses, observational research, developmental studies, and correlational
studies.
Questionnaire
The main justification for using a questionnaire is the need to obtain responses from
people, often from a wide geographical area. The questionnaire usually strives to secure
information about present practices, conditions, and demographic data. Occasionally, a
questionnaire asks for opinions or knowledge.
Interview
The interview and the questionnaire are essentially the same technique insofar as planning
and procedures are concerned. Obviously, the interview has certain advantages over the
questionnaire. The researcher can rephrase questions and ask additional ones to clarify
responses and secure results that are more valid. Becoming a skilled interviewer requires
training and experience. Telephone interviewing has become increasingly common in
recent years. It costs half as much as face-to-face interviews and can cover a wide
geographical area, which is generally a limitation in personal interviews. We discuss some
other advantages of the telephone interview technique in chapter 15.
Normative Survey
A number of notable normative surveys have been conducted in the fields of physical
activity and health. The normative survey generally seeks to gather performance or
knowledge data on a large sample from a population and to present the results in the form
of comparative standards, or norms. The AAHPER Youth Fitness Test Manual (American
Association for Health, Physical Education and Recreation, 1958) is an outstanding
example of a normative survey. Thousands of boys and girls ages 10 to 18 throughout the
United States were tested on a battery of motor fitness items. Percentiles were then
established for comparative performances to provide information for students, teachers,
administrators, and parents. The AAHPER youth fitness test was developed in response to
another survey, the Kraus-Weber test (Kraus & Hirschland, 1954), which revealed that
46
American children scored dramatically lower on a test battery of minimum muscular fitness
when compared with European children.
Case Study
The case study is used to provide detailed information about an individual (or institution,
community, and so on). It aims to determine unique characteristics about the subject or
condition. This descriptive research technique is used widely in such fields as medicine,
psychology, counseling, and sociology. The case study is also a technique used in qualitative
research.
Job Analysis
This type of research is a special form of case study. It is done to describe the nature of a
particular job, including the duties, responsibilities, and preparation required for success in
the job.
Observational Research
Developmental Studies
47
participants from different age groups to assess the effects of maturation. This is called the
cross-sectional approach.
Correlational Studies
Epidemiological Research
Another form of descriptive research that has become a viable approach to studying
problems dealing with health, fitness, and safety is the epidemiological research method.
This type of research pertains to the frequencies and distributions of health and disease
conditions among populations. Rate of occurrence is the basic concept in epidemiological
studies. The size of the population being studied is an important consideration in
examining the prevalence of such things as injuries, illnesses, or health conditions in a
specified at-risk population.
Although cause and effect cannot be established by incidence and prevalence data, a strong
inference of causation can often be made through association. Chapter 17 is devoted to
epidemiological research.
Experimental Research
Experimental research has a major advantage over other types of research in that the
researcher can manipulate treatments to cause things to happen (i.e., establish a cause-and-
effect situation). This approach contrasts with other types of research in which already
existing phenomena or data from the past are observed and analyzed. As an example of an
experimental study, assume that Virginia Reel, a dance teacher, hypothesizes that students
would learn more effectively through the use of video. First, she randomly assigns students
to two sections. One section is taught by the so-called traditional method (explanation,
demonstration, practice, and critique). The other section is taught in a similar manner
except that the students are filmed while practicing and can thus observe themselves as the
teacher critiques their performances. After 9 weeks, a panel of dance teachers evaluates both
48
sections. In this study, the teaching method is the independent variable, and dance
performance (skill) is the dependent variable. After the groups’ scores are compared
statistically, Virginia can conclude whether her hypothesis can be supported or not.
experimental research—A type of research that involves the manipulation of treatments in
an attempt to establish cause-and-effect relationships.
In experimental research, the researcher attempts to control all factors except the
experimental (or treatment) variable. If the extraneous factors can be controlled, then the
researcher can presume that the changes in the dependent variable are due to the
independent variable. Chapter 18 is devoted to experimental and quasi-experimental
research.
Qualitative Research
In the study of physical activity, qualitative research is the relatively new kid on the block,
although it has been used for many years in other fields, such as anthropology and
sociology. Researchers in education have been engaged in qualitative methods longer than
researchers in our field have. As previously mentioned, several names are given to this type
of research (ethnographic, naturalistic, interpretive, grounded, phenomenological,
subjective, and participant observational). Some are simply name differences, whereas some
have different approaches and points of focus. We have lumped them all under the heading
of qualitative research, because that term seems to be the most commonly used in our field.
49
and one qualitative.
50
Another random document with
no related content on Scribd:
The Project Gutenberg eBook of Die Cigarette
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.
Language: German
Stephan Dirk
Die Tabake werden in den Ballen, in denen sie aus dem Orient
ankommen, in den Fabrikationsgang gebracht. In der ersten Station
werden die Ballen geöffnet, aufgeteilt, die eng aneinander gepreßten
Blätter werden einzeln auseinander genommen und sortiert. Nach
der Sortierung werden die Blätter (nach ihren Provenienzen
geordnet) in große Holzkisten gefüllt und erwarten in dieser Form
den Mischungsvorgang. Die nächste Station ist der Mischungsplatz,
auf dem aus den Holzkisten in dem angegebenen Verhältnis der
Mischungsrezepte die Tabake der verschiedenen Provenienzen
schichtweise übereinander gelegt werden. Es ist dies ein ziemlich
großer Platz, auf den der Inhalt der Kisten gekippt und von Arbeitern
in gleichmäßigen Lagen auf der ganzen Fläche verteilt wird. Bei
diesem Prozeß wird der Tabak den jeweiligen unterschiedlichen
Anforderungen entsprechend mehr oder weniger angefeuchtet.
Dann wird das Gemisch in große Boxen gebracht, wo es einige Tage
lagert.
Die nächste Station ist die Tabakschneiderei. Das Tabakgemisch
wird großen Schneidemaschinen zugeführt, die die Blätter in feine
Strähnen zerschneiden. Der von den Messern herunterfallende Tabak
wird auf Transportbändern in eine Entstaubungsanlage gebracht, die
als nächste Station die Aufgabe hat, den Tabak von dem bitteren
Tabakstaub gründlich zu reinigen. Aus der Entstaubungstrommel
wird der Tabak wieder in große Holzkisten gefüllt und neuerlich
einem Zwischenlager zugeführt.