The Science of Attitudes, 1st Edition
Visit the link below to download the full version of this book:
https://medipdf.com/product/the-science-of-attitudes-1st-edition/
Click Download Now
First published 2016
by Routledge
711 Third Avenue, New York, NY 10017
and by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2016 Taylor & Francis
The right of Joel Cooper, Shane F. Blackman, and Kyle T. Keller to
be identified as authors of this work has been asserted by them in
accordance with sections 77 and 78 of the Copyright, Designs and
Patents Act 1988.
All rights reserved. No part of this book may be reprinted or
reproduced or utilised in any form or by any electronic, mechanical,
or other means, now known or hereafter invented, including
photocopying and recording, or in any information storage or
retrieval system, without permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks
or registered trademarks, and are used only for identification and
explanation without intent to infringe.
Library of Congress Cataloging-in-Publication Data
Cooper, Joel.
The science of attitudes / Joel Cooper, Shane Blackman, Kyle Keller.
pages cm
Includes bibliographical references and index.
1. Attitude (Psychology) 2. Attitude change. I. Blackman, Shane
J. II. Keller, Kyle. III. Title.
BF327.C665 2016
152.4—dc23
2015007593
ISBN: 978-1-138-82078-4 (hbk)
ISBN: 978-1-138-82079-1 (pbk)
ISBN: 978-1-315-71731-9 (ebk)
Typeset in Stone Serif
by Apex CoVantage
CONTENTS
Preface vi
CHAPTER 1 The Meaning and Measurement of Attitudes 1
CHAPTER 2 Attitude Strength and Structure 23
CHAPTER 3 The Functions of Attitudes 58
CHAPTER 4 Persuasion: Classic Approaches 83
CHAPTER 5 Dual-Process Theories of Attitude Change 114
CHAPTER 6 Predicting Behavior from Attitudes 145
CHAPTER 7 Predicting Attitudes from Behaviors 169
CHAPTER 8 Resistance to Persuasion 199
CHAPTER 9 Implicit Measurement of Attitudes 226
CHAPTER 10 ew Frontiers in Attitude Research:
N
Accessing and Modeling the Brain 258
References 283
Author Index 329
Subject Index 341
v
P R E FA C E
This book is about attitudes. Everyone has them. We have attitudes about important
world issues, about the mundane objects in our everyday life, and about the people
who comprise our social world. The attitude concept has been the most studied con-
cept in the history of social psychology. It is a topic covered in psychology courses,
from the basic introductory class to the graduate level. Yet, with all the attention
paid to attitudes, the current book was born of a teaching need. We found it difficult
to teach a responsible course on the social psychology of attitudes without compil-
ing a unique curriculum of ideas, references and concepts. This volume is designed
to provide students, teachers and scholars with the overarching framework that uni-
fies the exciting field of attitudes and persuasion and also brings to life the scientific
approach to the discipline.
When psychologists first began to study the attitude, they could only imagine
that some day researchers would be combining people’s verbal judgments with
information gathered from implicit measures and from brain activity. Now, modern
attitude research uses a panoply of traditional and innovative methods to form a
more complete picture of how people form and change attitudes. From its incep-
tion, the study of attitudes in psychology meant applying the scientific research
method. This book begins with an analysis of how attitudes first came to be studied
and shows the changing perspectives that occurred over the course of time. We
use a historical approach to trace the changing methods, theories and assumptions
that guided research in attitudes and attitude change. The book gives the reader a
foundation in the classic approaches that prepares the way for an appreciation of
current issues and cutting edge methodologies. We show the reader how contem-
porary issues complement classic approaches and how newer implicit measures and
fMRI technology complement more traditional measures in the field of attitudes.
No book can be written without the help of numerous people. We wish to
acknowledge and thank the many people who by reading, criticizing, commenting,
typing, taking photos and otherwise offering their assistance to allow us to com-
plete this book. We specifically acknowledge Paul Bree, Matthew Kugler, Christine
McCoy, Allison Fleming and Barbara Cooper for their valuable assistance. We also
could not have completed the book without the helpful criticisms and suggestions
offered by Russell Fazio, Michael Olson and other anonymous reviewers.
vi
CHAPTER 1
The Meaning and Measurement
of Attitudes
A young girl walked off her school bus, fearful that her school would close. She
was twelve years old, and nothing had been more important in her young life
than her education. Her hopes and dreams were threatened by an edict from the Tal-
iban militia: All schools serving girls must be shut and boarded by January of 2009.
As a child growing up in the Swat Valley of Pakistan, Malala Yousafzai’s schooling
had been frequently interrupted by the militias, hostile to the belief that girls might
receive a formal education. “How dare the Taliban take away my basic right to edu-
cation?” she had said to a public meeting of reporters when she was 11 years old.
Malala’s attitude toward education would cause her both trauma and triumph.
Trauma occurred on October 9, 2012, when Taliban gunmen boarded her school bus
and shot her with a single bullet to the head. They made clear that she had been
targeted for her outspoken commitment to girls’ education. But Malala survived
and, following a difficult recovery, emerged to a triumphant standing ovation at
the General Assembly of the United Nations. She concluded her remarks by saying,
Dear brothers and sisters, we want schools and education for every child’s bright
future . . . we must not forget that millions of people are suffering from poverty,
injustice and ignorance. We must not forget that millions of children are out of
schools. We must not forget that our sisters and brothers are waiting for a bright
peaceful future. So let us wage a global struggle against illiteracy, poverty and ter-
rorism and let us pick up our books and pens. They are our most powerful weapons.
One child, one teacher, one pen and one book can change the world. Education is
the only solution. Education First.
(NY Times, July 12, 2013)
Malala Yousafzai was awarded the Nobel Peace Prize in 2014.
Attitudes matter. Throughout the history of the world, people have taken
extraordinary steps to support a set of attitudes and beliefs to bring about a better
world. Gandhi, Nelson Mandela and Martin Luther King Jr. led societies to new views
1
2 The Meaning and Measurement of Attitudes
of human freedom and dignity by their written words and their behavior. Every day,
people take action to advocate for social justice. They persuade and organize in the
service of bringing about a world that is closer to the paragon in which they believe.
Attitudes matter in our daily lives as well. The way we think of our world is
largely determined by our attitudes. From the major political issues of the day to the
mundane evaluations of the food on our breakfast tables, it is difficult to envision
a world without them. They are reflections of our likes and dislikes, our preferences
and our evaluations. It seems virtually impossible to describe our social lives with-
out considering our attitudes. This book is devoted to organizing in a coherent fash-
ion the main tenets of what the science of psychology has learned about attitudes.
We will examine how our attitudes are structured, how they are measured and the
functions they serve. We will consider the effectiveness of ways to change attitudes
in the form of persuasion, and also the ways that allow people to resist persuasion.
We will consider the relationship of our attitudes to our behaviors, both as causes
and consequences of behavior, and finally discuss the emerging knowledge relating
attitudes to electrochemical changes in our brains.
A BRIEF HISTORY OF THE STUDY OF ATTITUDES
The scientific study of the attitude has a long and storied history. Although the
science of social psychology is only a little more than a century old, Thomas and
Znaniecki wrote in 1918 that social psychology was essentially the study of the
attitude. It was not always so. The first usage of the word attitude in the English
language was recorded in the 17th century, when it was adapted from the French.
Its meaning was not at all what psychologists describe today. If you referred to a
person’s attitude, you would be referring to their posture rather than a psycho-
logical concept. The first known link to psychology entered the language in 1725,
when attitude referred to a posture of the body that reflected a mental state. It was
not until 1837 that etymologists found a use of the word to refer to an opinion or a
feeling about an object. Advance the clock only 80 years and Thomas and Znaniecki
(1918) could declare the study of the attitude to be the essence of the fledgling sci-
ence of social psychology.
Although the term attitude to describe people’s mental states toward an object
or issue is relatively newly minted, the art of persuading people to change those
mental states is as old as civilization. It was the profession of the Sophists in ancient
Greece to teach citizens how to argue and persuade. Aristotle considered persuasion
in great detail in Rhetoric. The Roman forums relied on debate and persuasion in the
running of the empire. It is not that scholars were unaware that people held opin-
ions but, as art gave way to science, they needed a more precise way to conceptual-
ize people’s opinions and so the relatively new term, attitude, rose to prominence.
The Meaning and Measurement of Attitudes 3
The scientific study of attitudes was facilitated by the development of meth-
ods to assess them. L. L. Thurstone, Rensis Likert, Louis Guttman and Charles
Osgood provided increased sophistication to the scales used to study attitudes.
With the publication of attitude measurement techniques, the volume of empirical
research in attitudes exploded. By 1935, Gordon Allport declared the attitude to be
“the most distinctive and indispensable concept in contemporary American social
psychology” (Allport, 1935, p. 198). From the perspective of the 21st century, it is
clear that Allport’s assessment remains true, with the caveat that the qualifying term
“American” is no longer required. A literature search using “attitude” as the search
term yields more than 65,000 articles, chapters, books and dissertations.
Attitudes express our evaluations, influence our perceptions and guide our
behavior. A matter of practical and theoretical concern is how to alter them. Polit-
ical candidates know that people’s attitudes toward them, and possibly the issues
they represent, predict their voting behavior. Commercial marketers know that peo-
ple’s attitudes toward a product predict their purchasing behavior. Leaders of mass
movements know that people will sacrifice a great deal in support of attitudes they
find powerful and important. Finding the principles that govern the creation and
change of attitudes has been the focus of a healthy proportion of those 65,000
entries in the psychological literature.
A Tale of Two Traditions: Attitudes and Attitude Change
Persuasion is both an art and a science. As an art form, it relies on intuition. Many
political candidates, political consultants and advertising agencies are adept at con-
vincing their audiences without being able to articulate the principles that underlie
the persuasion. Just as artists paint a picture that can be striking for its beauty, some
persuasive communicators effectively convince their audience with the subtlety of
their words or the beauty of their images. When a young man named Volney Palmer
opened the very first advertising agency in Philadelphia in 1843, instinct and intui-
tion were the tools of his trade (Pratkanis & Aronson, 2001). The science of persua-
sion had not yet begun.
The science of persuasion, in contrast to the intuition of persuasion, relies on
the systematic investigation of underlying principles that lead to attitude change. The
first programmatic effort to examine attitude change systematically was conducted
by a group of investigators at Yale University in the 1940s. The group was con-
vened by the United States government, responding to what it perceived to be a
dilemma of public opinion. Toward the end of WWII, after the surrender of the
European Axis powers, the government was concerned about persuading its citizens
that the war in the Pacific theater of operations would be both lengthy and difficult.
Worried that people would believe that a successful end to the war was inevitable,
the government was determined to persuade them that sacrifice remained essential.
4 The Meaning and Measurement of Attitudes
They convened a group of social scientists, funded by the U.S. Army’s Information
and Education Division, to engage in the systematic scientific investigation into the
principles of persuasion.
The accomplishments of the Yale Attitude Change and Communication Pro-
gram were vast, and we will examine them in detail in Chapter 4. It is interesting
to observe, however, that the vast literature on persuasion gave little attention to
what was meant by an attitude. Investigators examined the principles of persua-
sion by using concepts such as opinions, beliefs and attitudes interchangeably. Hov-
land, Janis and Kelley’s (1953) major volume, reporting results of the early work,
was entitled Communication and persuasion: Psychological studies of opinion change.
The word attitude does not appear. It was as though an attitude was anything that
was measureable by a scale. Issues that served as targets for attitude change ranged
from whether people thought that Prohibition was a good policy to whether peo-
ple believed the Soviet Union could build a nuclear submarine. The application of
scientific methodology to studies of persuasion and the growth of new theories to
predict change fueled a long and exciting period of study in attitude change. Being
certain of what we mean by an attitude has had a more deliberate growth.
Why Study Attitudes?
Social scientists study attitudes because they are psychological structures that pro-
vide insight into human thought and behavior. Understanding the principles that
underlie attitudes and attitude change tells us a great deal about human develop-
ment, learning and motivation. We also study attitudes because they are a bridge
to human behavior. It is reasonable to believe that people’s behavior is affected by
their attitudes. Therefore, understanding attitudes provides insights into people’s
behavior. Understanding how attitudes change can provide insight for changing
behavior.
Imagine a candidate for political office who hopes to persuade people to vote
for her. By changing people’s attitudes toward the issues she stands for, it seems rea-
sonable that people will be more likely to cast their ballot for her. Similarly, a busi-
ness that hopes to improve its profits may hope to persuade people to adopt positive
attitudes toward the product it manufactures. As a practical matter, being able to
persuade people to adopt and/or change their attitudes is an important enterprise
and a scientific approach to understanding the principles that underlie persuasion
facilitates that enterprise.
The attention that social scientists have paid to the study of attitudes has pro-
duced important theories and results. Yet anomalies in the findings have raised
questions about the basic assumptions underlying the attitude concept. For exam-
ple, attitudes do not always predict behavior. In one important illustration that we
shall consider in more detail in Chapter 6, LaPiere (1934) asked shopkeepers about
The Meaning and Measurement of Attitudes 5
their attitudes toward serving Chinese couples in their places of business. This was
an era in which anti-Chinese sentiment was strong in the United States. Almost
unanimously, shopkeepers expressed negative attitudes toward serving Chinese
couples and reported their intention to refuse them service. Yet LaPiere escorted a
Chinese couple on an automobile trip covering more than 10,000 miles, stopping at
251 hotels and restaurants. He found that, despite the attitudes that the proprietors
reported on the questionnaires, 250 of the 251 establishments gave full and polite
service to the couple.
Anomalies can be useful because they lead to more focused attention on resolv-
ing them. Following LaPiere’s report, many social psychologists turned their atten-
tion to studying the structure, meaning and function of attitudes. This attitude
tradition was more precise than the attitude change literature in identifying some
of the underlying concepts that make attitudes more or less predictive of behavior.
Because all attitudes are not created equal, asking people for their attitude about a
cereal, a candidate or tooth brushing on an attitude scale may miss some important
differences among those attitudes. For example, some attitudes are stronger than
others and have greater impact on behavior. Attitude strength, in turn, depends on
such questions as how the attitudes were formed, how easily they can be activated
from memory and the consistency of the various components that comprise the
attitude. We will discuss these issues in the following chapter and then consider the
functions that they serve in Chapter 3. We will now turn our attention to consider-
ing what we mean by an attitude.
Defining the Attitude
Social psychology has made enormous progress in understanding the underlying
structure and function of attitudes, as well as developing empirically based theo-
retical perspectives on how to change them. There has been less consensus about
how to define them. We will adapt a definition of attitudes proposed by Zanna and
Rempel (1988):
An attitude is the categorization of a stimulus object along an evaluative dimension.
Zanna and Rempel went on to propose that the evaluation itself is a result of affec-
tive information, cognitive information, information about behavior, or some com-
bination of the three.
The critical notion in the definition of attitudes is evaluation. When we evalu-
ate a stimulus object, we are making a summary judgment about its value or worth
along a positive/negative dimension. This can be done as an absolute judgment,
such as “I think my elected official is good” or it can be a relative judgment, such
as “I think my elected official is better than my neighboring elected official.” As
6 The Meaning and Measurement of Attitudes
the number of categories increase from merely good versus bad to express grada-
tions of good and bad, we then have an ordinal scale for expressing our evaluative
judgments.
Although most scholars agree that attitudes are evaluations, they do not always
agree on what is meant by evaluation. One widely cited definition of attitudes was
placed into the literature by L. L. Thurstone (1946). He declared that attitudes were
“the intensity of positive or negative affect for or against a psychological object”
(p. 39). Note that the difference between the two views is that our proposed defi-
nition uses the term evaluation while Thurstone used the term affect. We construe
affect to be an emotive experience: When you feel happy, you experience affect;
when you feel sad, you experience affect. Although your affect may be one basis
for your evaluation of an attitude object, it may not be the only one. Consider ice
cream: Merely looking at a cold, creamy dish of chocolate ice cream may make you
experience delight. Without question, you feel happy and delighted. At the same
time, you know that ice cream will not help you lose the five pounds you have been
trying to lose, and that the fat and sugar content will not be good for your health. So
what is your attitude toward the ice cream? Your affective response is certainly pos-
itive, but is tempered with what you believe about its health consequences. Asked
about your attitude toward ice cream, your evaluation will be a complex judgment
based upon a combination of the factors you have considered.
Other scholars have proposed that affect, cognition and behavior are all
intrinsic aspects of the attitude concept (Rosenberg & Hovland, 1960; Zimbardo &
Ebbesen, 1970). In this tripartite view, an attitude must include a tendency for a
person to act in a way that corresponds to his or her cognitive beliefs and affective
experiences. For example, Droba (1933) suggested “an attitude is a mental disposi-
tion of the human individual to act toward or against something in the environ-
ment” (p. 309). In this view, your attitude about chocolate ice cream is positive to
the extent that your emotional experience, your beliefs about the ice cream and
your behavior toward it are all positive. Insisting on the behavioral tendency as
part of the definition of attitude raises the following question: If you know that ice
cream makes you feel good but you never eat it, is it legitimate to say that you have
a positive attitude toward ice cream?
We believe the answer to that question is yes. We concur with many attitude
researchers (e.g., Alberracin, Zanna, Johnson and Kumkale, 2005; Zanna & Rempel,
1988) who suggest that to include behavior in the definition of attitudes pre-judges
some of the most fascinating and important questions in attitude research: Under
what circumstances do attitudes predict behaviors? Under what circumstances do
behaviors predict attitudes? Under what circumstances are behaviors predicted by
purely emotional experiences independent of any logical thought? These are empir-
ical questions whose answers have shed light on many fascinating issues. As we
present some of the classic research on attitudes in the forthcoming chapters, we
The Meaning and Measurement of Attitudes 7
will see that people’s evaluative judgments are sometimes based upon their behavior
toward an attitude object, sometimes on their thoughts and sometimes on their emo-
tive experiences—and, most often, on some weighted combination of the three. To
constrain attitudes by defining them as necessarily based on all three components
acting in unison is to eliminate some of the most interesting research questions
about attitudes. Using a working definition of attitudes as a summary evaluation
that takes into account any or all of the three sources of information (affect, behav-
ior and cognition) affords the most flexibility in studying attitudes.
The definition of attitude we are adopting also takes no position about the
stability or permanence of an attitude. An attitude may or may not be stored in
memory, ready to be used when interacting with an attitude object. A person may
have an often-used, well-rehearsed attitude toward chocolate ice cream that serves
her well whenever she approaches it. She sees it and she evaluates it positively with
no hesitation. Another person may have to consider each interaction with the ice
cream anew, depending on how she feels at the moment, the time of day, her weight
loss goals and so forth. Presumably, attitudes that are stored in memory may be
more stable, predict behavior more consistently and lead to quicker reactions about
the object than attitudes that are inferred at the moment (Fazio, 2001). However,
this is a research question rather than a definitional question and we shall examine
some of that research in the next chapter.
Attitude Measurement
Attitude assessment in the 21st century is a multi-billion dollar business. Preceding
any election, opinion polls abound. Those polls measure people’s attitudes on a
weekly or even daily basis. They may be conducted by personal interviewers, by
phone, by mail or by internet. Following televised debates, attitudes can be assessed
within minutes of the debates. The profession of attitude assessment is exemplified
in marketing and consulting firms whose expertise is to assess people’s attitudes
toward commercial products and advertisements, as well as political issues and
candidates.
Measurement is not a simple issue. Despite the sophistication of polling and
attitude assessment techniques, more than one candidate has been misled because
voters’ attitudes were assessed poorly. In a famous illustration, Republican candi-
date Thomas E. Dewey was advised to stop campaigning weeks before the 1948
U.S. presidential election because polling data had shown that an overwhelming
majority supported his candidacy. Worried that the only thing he could accomplish
by continuing to campaign was to stumble, he left for Europe with weeks remaining
in the campaign. On the day of the election, November 2, 1948, Harry S. Truman
became the 33rd president of the United States, upsetting Dewey by more than two
million votes. It may never be known if Truman ‘came from behind’ during the
8 The Meaning and Measurement of Attitudes
final weeks of the campaign or if the polling agencies simply measured the voters’
attitudes incorrectly.
The Stimulus Object
How one assesses attitudes is partly a function of what you want to know. We assess
people’s evaluation of a stimulus object, by which we mean almost anything
about which people can form evaluations. A stimulus object (sometimes called the
attitude object) can be a person, an animate or inanimate object or a topical issue. If
people can evaluate the stimulus object on a good/bad dimension, then it can serve
as the basis for an attitude. One of the primary concerns of attitude researchers is
to identify the stimulus object properly. Put another way, it is imperative to decide
what it is that you really want to know.
Imagine that you want to run for political office. Your supporters tell you that
you have a certain charm or charisma, but you also want to be certain that your own
attitudes about local and national issues are consistent with those of your electorate.
In this hypothetical scenario, you decide that expanding preschool opportunities
for economically disadvantaged children is your top priority and you want to assess
the attitudes of your potential constituents. You decide to conduct a poll in which
you will ask people about their attitudes. Before you can construct a scale, you need
to make an important decision: What is the stimulus (i.e., attitude) object? Do you
want to know whether people are in favor of preschool education in general, or
do you want to know whether they favor a specific proposal currently before their
legislature? Do you want to know if they favor public funding of preschools or just
having a system that might be funded at some time in the future? Or, perhaps you
want to know how favorable your potential constituents are toward helping disad-
vantaged children as a category of people. You may well find different evaluations
or attitudes, depending on how you construe the attitude object.
In the pages that follow, we will discuss a number of techniques that researchers
have frequently used to assess attitudes. Here we pause to note a scientific imper-
ative: Questions must be asked in a way that does not ‘push’ the answer to the
question. For example, if a respondent answers a question such as, “Do you agree
with a social policy that takes care of children, or would you rather have them grow
up without an education?” you are likely to find what you might mistakenly con-
strue as attitudinal consensus for preschools. However, the phrasing of the question
may have pushed people toward that answer because they do not want to see chil-
dren wandering the streets without an education. That nonetheless may be quite
different from a positive evaluation of a particular education assistance program
currently before the legislature.
In science, researchers must have no vested interest in the outcome of an atti-
tude questionnaire. If they phrase a question that pushes or leads people toward a
particular answer, it will only serve to weaken the meaningfulness of results they
obtain in their research. Sometimes, however, ‘push’ or leading questions are used
The Meaning and Measurement of Attitudes 9
deliberately for politicians or pundits to make a particular point. The wording of
questions can push people to appear to support a pundit’s particular viewpoint or
make it appear that a particular candidate is favored by a greater proportion of the
electorate than is really the case. The science of attitude research strives to make
certain that attitudes are measured, not pushed.
Explicit Attitude Assessments
Measurement is the assignment of numbers to objects according to rules (Stevens,
1946). Typically, scientific research assesses attitudes by measuring and assigning
numbers so that comparisons can be made. A person may be relatively favorable
to expanded preschool education for children. We may discover this by asking
an explicit question that assigns a number that can be used to compare against a
standard. For example, a person’s attitude measurement may be greater than the
mid-point on a favorable/unfavorable dimension or it may be greater than the num-
ber that other people marked on the favorable/unfavorable dimension.
How do we obtain these numbers, and how certain are we that the numbers
represent people’s evaluations? Arguably, the dawn of the age of scientific attitude
research began with L. L. Thurstone’s classic article whose main thesis (and title) was
the pronouncement that “attitudes can be measured” (Thurstone, 1928, p. 529). It
seems ironic from the perspective of modern psychology that there was a time that
attitudes were thought to be unmeasurable. But that was the case in the early part
of the 20th century. Thurstone argued that, contrary to the prevailing assumption,
numbers could be assigned to attitudes in a scientifically valid manner. He likened
the problem of measuring a multi-faceted concept like an attitude to measuring a
multi-dimensional object like a table.
. . . Measuring an ordinary table is a complex affair, which cannot be wholly
described by any single numerical index. So is a (person) such a complexity, which
cannot be wholly represented by a single index. Nevertheless, we do not hesitate to
say that we measure the table.
(Thurstone, 1928, p. 531)
The Thurstone Scale
Thurstone advanced a technique to assess attitudes that now bears his name. He rea-
soned that people’s attitudes are given expression in the form of specific opinions
that they would endorse about a stimulus object. To use a modern example, peo-
ple who endorse an opinion statement like “All children should have access to an
education when they are young” are likely to have attitudes that are supportive of
publicly funded preschools. People who endorse statements like “Children should
begin formal schooling at age 6” are more likely to have negative attitudes about a
public preschool program.
10 The Meaning and Measurement of Attitudes
Constructing a Thurstone Scale requires several steps. The first is to ask a large
number of people to generate opinion statements like the previous ones. Further
examples might be “Preschool for all children is too expensive,” and “Preschool for
all children is a moral responsibility of a society.” A researcher can gather hundreds
of statements, but the goal of Thurstone’s method is to turn them into a scale. This
requires that there be a scale of measurement such that people’s attitudes can be
legitimately compared. To create the scale, the opinion items that were generated by
the first large group of respondents are shown to a different group of respondents,
who are asked to separate the items into 11 piles. One pile will contain all of the
opinion items that they judge to be extremely supportive of preschools for the disad-
vantaged, another pile will contain all of the items that are extremely anti-preschool,
and yet another pile will be neutral, neither favoring nor opposing preschools for
the disadvantaged. Therefore, based on the judges’ ratings, there will be 5 units of
pro-preschool that vary in intensity with equal intervals from extreme to moderate
and five of anti-preschool also ranging from extreme to moderate and one neutral
point. The important operation that allows the scale to have units of measurement
is the judges’ separation of the opinion statements into the 11 equal units.
A Thurstone scale can be created for any topic on which people can make
evaluative judgments. Respondents simply indicate which of the statements they
endorse, with the mean of the endorsed statements constituting a numerical rep-
resentation of the respondent’s attitude. The scale in Figure 1.1 illustrates a hypo-
thetical Thurstone scale that may have been constructed for determining people’s
PLEASE INDICATE WHETHER YOU AGREE OR DISAGREE WITH EACH ITEM
1. Free preschools for the poor are socialist
2. The Head-Start preschool program has been a failure
3. Taxes are too high to support preschools for everyone
4. People should pay for their own children’s preschool
5. Nations that have preschools have poor education systems
6. Children can get by without preschools
7. In the long run, we all pay for children who are uneducated
8. People cannot be sure that public education will always be available
9. My child needs a free preschool program
10. Poor children can only succeed if preschools are available
11. Providing education from birth to adulthood is a moral obligation for
society
FIGURE 1.1 Example of a Thurstone scale for measuring attitudes towards
preschool education. Participants respond by indicating which statements
they endorse.
The Meaning and Measurement of Attitudes 11
attitudes toward preschool for disadvantaged children. If John decides that he
endorses items numbers 6, 7 and 8, we can assign him a number on the scale that is
the mean (=7) of the items he endorsed. That number can be used to test differences
among individuals or among groups of people who take the scale.
Thurstone scaling is an important technique in attitude measurement, partially
because of its historical importance. As we pointed out, social scientists were eager
to study people’s attitudes but, until Thurstone’s paper, they did not have a psycho-
metrically reliable method for studying them. A second reason for the importance
of Thurstone scales is that they could be developed to study attitudes about any atti-
tude object. Thurstone scaling is still used in modern social psychology, although
not to the degree that the scales that succeeded it are used.
The Likert Scale
One drawback of constructing an attitude scale using Thurstone’s procedure is the
cumbersome nature of the approach. By drawing on hundreds of people to generate
a list of statements, Thurstone hoped to avoid having an unrepresentative set of
statements to rate. By recruiting another hundred or so judges to rate the statements
for their level of support for a stimulus object, he hoped to create consensual units
for the scale that respondents could endorse. The burden on researchers each time
they wished to study an issue seemed nearly prohibitive.
Rensis Likert (1932) suggested a simpler method. Not only was he seeking a scale
that would be easier to use, but he also was concerned with Thurstone’s assumption that
asking a large number of judges to rate a large number of statements solved the poten-
tial problem of having biased or unrepresentative issues. His simpler method involved
generating ‘judgments’ with which people could agree or disagree at various strengths.
A Likert item on pre-school education would look like the following:
“A modern democracy should have a system of free preschools for disadvantaged children.”
1 2 3 4 5
Strongly Undecided Strongly
Agree Disagree
FIGURE 1.2 Example Likert scale for measuring attitudes towards
preschool education. Participants indicate their response by circling a
number on the scale.
Respondents are asked to choose the degree to which they agree or disagree
with the item using a 5-point scale like the one pictured in Figure 1.2.
Note the differences between Likert’s and Thurstone’s methods. Thurstone
asked people to endorse statements of beliefs toward an attitude object that had been
pre-rated by a group of judges to fall equally along a value continuum. Likert asked
people the degree to which they agreed or disagreed with a judgment that contained
