Description
Your assignment is to write a short (~3
pages, or 900 words) APA-style paper in which you discuss a pressing social problem,
summarize a social psychological theory, and describe how this theory could be (or has been)
applied to help understand or solve this problem. These papers are worth 50 points each.
Start your paper by describing a pressing social problem (~1 page) and provide at least one
reference. In summarizing the nature of the problem, make sure to link the cause of the
problem to human behavior. I suggest being as specific as possible. For example, rather than
talking about “environmental problems”, it’s better to focus on just one problem such as the
availability of fresh water, and even better, the availability of fresh drinking water in California.
Make sure to link the problem to human behavior, for example the large amount of fresh water
that is used for outdoor irrigation, such as watering grass.
As a starting point, here’s a short list of possible problems, but feel free to choose others that
are not included in the list below:
• Environmental problems, such as climate change, species extinction, depletion of
natural materials, water pollution, overpopulation, …
• Poverty and economic disparities, prejudice, discrimination, obesity, homelessness
• Educational disparities, unemployment, racial and gender inequality
• Violence and aggression, bullying
• Conflicts in relationships, divorce, child abuse, sex trafficking, dating violence
• Substance use and abuse, vaping, smoking, binge drinking, impaired driving
Second, discuss a social psychology theory or principle (~1 page). You’re free to draw on any
theory or principle discussed in class, or in your course readings. Make sure to describe the
details of the theory, and include at least one reference. Examples of social psychological
theories from this section of the course include: stereotypes, in-group bias, prejudice, contact
hypothesis, sociobiology, attraction, triangular theory of love, intimacy, disclosure reciprocity,
decision-making model of prosocial behavior, pluralistic ignorance, diffusion of responsibility,
frustration and aggression, weapons effect, social learning theory, and catharsis.
Finally, write a section in which you use the theory or principle to either understand or solve
the problem (~1 page). It’s also appropriate here to describe a previous attempt to address the
problem, and discuss whether it was effective or ineffective. Include at least one reference.CHAPTER 6
ATTITUDES AND BEHAVIOR
Attitudes
ABCs and the Bases of Attitudes
Attitudes as Cognitive Networks
Attitude Construction
Measuring Attitudes
Explicit Measures
Goal-Directed Behavior
Implicit Association Test
Social Neuroscience
Origins of Attitudes
Mere Exposure
Classical Conditioning
Operant Conditioning
Observational Learning—Imitation and
Modeling
Impressionable Years
Do Attitudes Predict Behavior?
Research on Explicit Measures of
Attitudes
Implicit Attitudes and Behaviors
Expectancy-Value Theory
The Theory of Planned Behavior
Behaviors Predict Attitudes
Cognitive Dissonance Theory
Self-Perception Theory
Summary of the Relationship Between
Attitudes and Behaviors
Summary
Key Concepts
Review Questions
AQ: Quote
text with
author name
is missing
183
Woo00037_Ch06.indd 183
05-01-2023 18:57:47
184
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
Attitudes
The concept of attitudes is one of the oldest and most important in the field of social psychology. In short, attitudes are our likes and dislikes (Schwarz & Lee, 2019).
An attitude can be about anything: a person, a group, an event, or even an abstract
concept such as climate change or social media. More formally, attitudes are favorable or unfavorable evaluations of an attitude object (Albarracín et al., 2019;
Oskamp & Schultz, 2005). Given this definition, it’s clear that attitudes extend to
almost all aspects of our lives. Attitudes about another person can serve as the basis
for interpersonal attraction and relationships; attitudes toward oneself are the basis for self-esteem; attitudes toward products can influence our consumer choices;
and attitudes toward academic majors can influence student course selection and,
eventually, career pathways.
Although we hold attitudes on a wide range of topics, they are clearly evident in
the context of public opinion. Public opinion refers to the attitudes and beliefs held
by a group of individuals about a specific topic. Public opinion is often measured
through polls, and it is used extensively in politics and elections. It reflects majority
beliefs and is used as a tool for measuring conflicting group interests (Glynn et al.,
2016). Consider the following attitudes, as measured with public opinion polls
conducted by the Gallup organization using nationally representative samples of
people living in the United States (cf., Gallup.com, 2022):
Woo00037_Ch06.indd 184
•
Tobacco products. “In general, how harmful do you feel each of the following is to the people who use them: e-cigarettes, also known as ‘vaping’;
chewing tobacco; cigarettes?” Response: Thirty-eight percent reported
e-cigarettes as very harmful, 71% reported chewing tobacco as very harmful, and 82% reported cigarettes as very harmful.
•
Smoking. “Should smoking in this country be made totally illegal or not?” Response: Twenty-five percent responded that smoking should be made illegal.
05-01-2023 18:57:47
CHAPTER 6: Attitudes and Behavior
185
•
Marijuana. “Do you think the use of marijuana should be made legal or not?”
Response: Sixty-six percent responded that marijuana should be made legal.
•
Prejudice. “Thinking honestly about your feelings, how much prejudice,
if any, do you feel toward [Muslims as a religious group]?” Response:
Fifty-seven percent reported “none,” 14% reported “a little,” 20% reported
“some,” and 9% reported “a great deal.”
•
Same-sex marriage. “Do you think marriages between same-sex couples
should or should not be recognized by the law as valid, with the same
rights as traditional marriages?” Response: Sixty-three percent reported
that same-sex marriages should be valid.
•
Moral acceptability. The survey asked respondents for their personal beliefs
regarding whether the following are morally acceptable or morally wrong
(see Figure 6-1 for the entire list of 21 social issues):
•
•
Birth control—92% reported as morally acceptable.
Gambling—68% reported as morally acceptable.
Americans’ views of the moral acceptability of 21 issues
Regardless of whether or not you think it should be legal, for each one, please tell me whether
you personally believe that in general it is morally acceptable or morally wrong.
92
% Morally acceptable
89
% Morally wrong
79
85
77
71
68
65
64
64
17
58
45
31
18
61
53
34
79
66
63
60
33
80
35
52
51
44
44
50
42
38
35
37
31
31
28
19
20
12
9
Woo00037_Ch06.indd 185
Cloning humans
Married men and women
having an affair
Suicide
Polygamy
Cloning animals
Pornography
Sex between teenagers
Abortion
Medical testing on animals
Doctor-assisted suicide
The death penalty
Buying and wearing clothing
made of animal fur
Gay or lesbian relations
Having a baby outside
of marriage
Embryonic stem cell research
Gambling
Smoking marijuana
Divorce
Sex between an unmarried
man and woman
Drinking alcohol
Birth control
6
FIGURE 6-1
Americans’ views
of the moral
acceptability of
21 issues.
05-01-2023 18:57:48
186
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
•
•
•
•
•
•
Gay or lesbian relations: 63% reported as morally acceptable.
Buying and wearing clothing made of animal fur: 53% reported as
morally acceptable.
Sex between teenagers: 38% reported as morally acceptable.
Pornography: 37% reported as morally acceptable.
Cloning humans: 12% reported as morally acceptable.
Married men and women having an affair: 9% reported as morally
acceptable.
ABCS AND THE BASES OF ATTITUDES
Cognitive
dimension: The
cognitive dimension
of an attitude involves
beliefs about the
attitude object.
Affective
dimension: The
affective dimension
of an attitude refers
to the feelings and
emotions evoked by
the attitude object.
Behavioral
dimension: The
behavioral dimension
consists of a readiness
to respond in specific
ways to the attitude
object.
Research on attitudes has focused on three dimensions: cognitive, affective, and
behavioral (See Figure 6-2). The cognitive dimension of an attitude involves beliefs about the attitude object. For example, let’s take the case of a person’s attitude
toward electric cars. As a starting point, the person holds a range of beliefs about
electric cars: they accelerate quickly, they are cheap to drive, they have a limited
driving range, and they are very quiet. The affective dimension of an attitude refers
to the feelings and emotions evoked by the attitude object. In our example of electric cars, this is the degree to which the person holds positive or negative appraisals
of the features of electric cars. So, in the previous list, the positive and negative appraisals are as follows: they accelerate quickly (which is thrilling), they don’t use gas
(which is socially responsible), they are quiet (which brings social approval from
others), and they have limited driving range (which evokes a sense of fear of getting
stranded). Finally, the behavioral dimension consists of a readiness to respond
in specific ways to the attitude object. In our example of an electric car, has the
individual ever driven an electric car? Does the individual intend to purchase an
electric car in the future? Think of these three dimensions as the ABCs of attitudes:
affective, behavioral, cognitive.
Behavioral
“I am so getting
one of these!”
Affective
“Yay! So fun to
drive”
“Yikes! What if I
get stranded?”
Cognitive
“So quiet”
“No more
buying gas!”
FIGURE 6-2
The three dimensions: of attitudes: cognitive,
affective, and
behavioral
Woo00037_Ch06.indd 186
05-01-2023 18:57:49
CHAPTER 6: Attitudes and Behavior
187
Attitudes are the collection of these affective, behavioral, and cognitive dimensions (Fabrigar et al., 2019). Continuing with our example of our attitudes toward
electric cars, my attitude is fairly positive. Although I have limited behavioral experience with an electric car (I’ve never driven one), I do have favorable beliefs and
primarily positive feelings toward them. Importantly, each of these dimensions can
serve as the basis for a person’s attitude, and some may be weighted more heavily
than others (Crites et al., 1994). Although my attitude toward electric cars tends to
be favorable, my wife has a more negative view. She shares my positive cognitions
about the quick acceleration, lack of gas use, and quiet driving, but for her, the fear
associated with getting stranded because of the limited range is extremely high, and
as a result, she has a more negative attitude. In her case, the affective dimension
serves as her primary basis (and as a result, she holds a negative attitude), whereas
my attitude is more cognitively based (and as a result, my attitude is less affected
by the fear associated with limited range). Note that each of the dimensions of an
attitude can serve as the underlying basis, each to varying degrees.
ATTITUDES AS COGNITIVE NETWORKS
Dating back to the 1970s, basic psychological science has focused heavily on how
the brain stores, organizes, and processes information. This research has also been
strongly influential in social psychology, most notably in the research on social
cognition that was previously summarized in Chapters 3, 4, and 5. This social cognitive perspective has helped to advance our understanding of how individuals
make decisions, understand themselves, and perceive others. It has also strongly
influenced our understanding of attitudes.
From a social cognitive perspective, attitudes are extensions of the brain’s way
of organizing information (Dalege et al., 2016). But the critical extension is that
attitudes involve evaluations—that is, attitudes are more than just associations in
our semantic network; they also include the affective, behavioral, and cognitive
evaluations that we discussed previously as the bases for attitudes. As an example,
let’s consider a person’s attitude toward a presidential candidate. This example was
described by Dalege at al. (2016) regarding Bob (the person) and President Obama
(the attitude object), but you can extend it to any attitudinal object. The hypothetical attitude network is shown in Figure 6-3, along with the formation and growth of
Bob’s attitude over time (from panel A to panel D). To start, Bob uses information
from a few social media posts and video clips to form an initial impression: Obama
is caring, honest, and moral (panel A, shown in red). So, at the start, Bob has a weak
favorable attitude toward President Obama based on these beliefs. As Bob thinks
about it more, his attitude network grows with the additional set of cognitive evaluations shown in blue: competency, leadership, and intelligence. At this point, the
cognitive associations trigger affective evaluations, and Bob feels a sense of pride,
hope, and sympathy for Obama (shown in green). And finally, Bob starts to take
action: he puts a bumper sticker on his car supporting Obama, he tries to convince
his friends to support Obama, and he votes for Obama in the primary election
(shown in yellow).
Woo00037_Ch06.indd 187
05-01-2023 18:57:49
188
FIGURE 6-3
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
A
B
An example of the
formation of an attitude network.
Note: These illustrations show a hypothetical attitude network as
it develops across time
in relation to a person’s
attitude toward a U.S.
presidential candidate.
In the first panel (a),
the attitude is newly
formed and consists
of three beliefs (caring, honest, moral).
The second panel (b)
shows the continued
growth of the attitude
network, including
additional evaluative
beliefs about competency, leadership,
and intelligence. The
third panel (c) shows
additional affective
evaluations, including
feelings of pride, hope,
and sympathy. Finally,
the fourth panel (d)
shows the addition of
behavioral enactments,
including convincing
others to vote for the
candidate, posting a
bumper sticker on their
car, and voting for the
candidate in a primary
election.
Source: Dalege et al.
(2016, p. 6).
Woo00037_Ch06.indd 188
moral
cares
honest
honest
cares
moral
leader
competent
intelligent
C
D
proud
good
competent
sympathy
leader
cares
hope
honest
intelligent
honest
moral
like
moral
hope
intelligent
leader
competent
vote
cares
sticker
proud
sympathy
convince
The cognitive-network framework provides a useful tool for understanding
attitudes. The model just described and shown in Figure 6-3 is from research
on causal attitude networks (CANs), one of several network models used to
represent attitudes (Dalege et al., 2017). In CAN models, the specific items are
referred to as nodes, and the links between the nodes are referred to as edges.
In our example, all the nodes are positive in valence, but the CAN model also
allows for negative connections. The CAN model offers a useful guide for understanding how attitudes form, their structure, and how they might change
over time. Here we’ll highlight just a couple of features of the CAN model. First,
note that the structure of an attitude includes the three bases we introduced previously: cognitive (shown in red and blue), affective (shown in green), and behavioral (shown in grey). Attitude is the aggregation of these three dimensions.
Second, attitude networks strive for consistency, meaning that if the connection
between two nodes is positive, then neighboring nodes are pressured to also be
positive. And finally, individuals are motivated to hold accurate attitudes. This
motivation for accuracy means that in some instances, positive and negative
nodes may be linked.
05-01-2023 18:57:49
CHAPTER 6: Attitudes and Behavior
189
Within social psychology, network models, like the CAN model, have been
highly influential. In later sections of this chapter, we’ll explore some of the measurement techniques that have been developed to assess these evaluative associations, as
well as research findings regarding the relationship between these associations and
a person’s behavior. But for now, let’s consider one of the important implications of
these cognitive networks. Intuitively, we might think of attitudes as specific objects
that reside in a person’s head. But from a network perspective, attitudes are derived
from the various associations, and they don’t exist independently. A cognitive network is constantly changing, and depending on the social context, some nodes will
be more accessible than others. Each node in the network has connections to other
nodes, and the activation of a node spreads to other nodes that are linked. In this
network perspective, attitudes can also exist independently of our conscious awareness, and in fact, we can even hold attitudes that are inconsistent with our personal
values (cf., Petty et al., 2012). For example, consider the person who values the
importance of educational opportunities for immigrants to their country but yet
also has nonconscious prejudicial attitudes toward specific immigrant groups. We
will examine unconscious attitudes in greater depth in Chapter 8 (Implicit Biases).
ATTITUDE CONSTRUCTION
Traditionally, attitudes have been viewed as relatively stable across time and context. The underlying assumption is that attitudes exist in a person’s head, like a
memory, and that they are recalled as needed in different situations. This dispositional approach views attitudes as something that a person “has” and that influences their actions. However, as summarized earlier, research on the structure
of attitudes has not supported this approach. Instead, research suggests that attitudes are more fluid and are strongly influenced by their level of activation and
the social context. Attitude construction views attitudes as evaluative judgments
that are formed on the spot, rather than stable dispositions that a person holds
(Schwarz, 2007).
To illustrate, consider your attitude toward Donald Trump using an 11-point
scale ranging from 0 = don’t think highly of him to 10 = think very highly of him.
Let’s say that your answer was 3—moderate but low. But what would happen if the
11-point scale instead ranged from –5 to +5? From a dispositional perspective, the
scores should be comparable, albeit adjusted downward to the new scale (so your
expected score would be –2). But in fact, in studies comparing attitudes toward political candidates, the scale ranging from –5 to +5 generally produces more favorable
attitudes than does the 0–10 scale (Schwarz & Bohner, 2001). The interpretation,
from an attitude-construction perspective, is that the use of negative options in the
rating scale involves comparing the candidate to negative associations, whereas the
use of the 0–10 scale simply involves the absence of positive thoughts. That is, when
constructing an attitude, the evaluative judgment changes if the person thinks of
negative associations (which some people hold about Donald Trump) versus the
absence of positive associations (which many people hold about Donald Trump).
Woo00037_Ch06.indd 189
Attitude
construction: views
attitudes as evaluative
judgments that are
formed on the spot.
05-01-2023 18:57:49
190
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
Measuring Attitudes
Like many constructs in psychology, attitudes are hypothetical. You can’t directly
observe an attitude, so instead, we infer them from what a person says and does.
In social psychology, a considerable amount of research has focused on measurement issues, and many techniques have been developed for inferring a person’s
attitude (Krosnick et al., 2019). These include explicit self-report measures, such
as the Likert scale or the feeling thermometer, along with a growing set of indirect
implicit measures, such as cognitive tasks and those that monitor brain activity.
EXPLICIT MEASURES
By far, the most widely used approach to measuring attitudes involves direct
self-report. Explicit measures involve asking the person to respond to questions
about the attitude object. Open-ended questions provide respondents with a free
choice about what to mention and how to answer. An example of an open-ended
question is, “What do you think is the most serious issue facing our country today?”
In contrast, closed-ended questions present two or more alternative answers for the
person to choose between—for example, “Do you think the recreational use of marijuana should be made legal or not?” Because social psychology is primarily quantitative in methodology (see Chapter 2 – The Science of Social Psychology), attitudes are
typically scored on a numeric scale that ranges from unfavorable to favorable ratings.
Although many scaling techniques have been used in the study of attitudes, the
Likert scale is by far the most common. The scale gets its name from its creator,
Rensis Likert (pronounced līk-ärt), who first published the technique as part of
his doctoral dissertation work in 1932. Likert’s method measures the intensity of a
person’s agreement with a question, rather than simply using a yes–no format. In
the Likert method, respondents are presented with attitudinal statements about a
given topic and are asked to rate each on a 5-point scale that ranges from strongly
agree to strongly disagree. Consider the following question about patriotism (taken
from Likert’s original 1932 publication).
We should be willing to fight for our country whether it is in the right or in the wrong:
_____ Strongly disapprove
_____ Disapprove
_____ Undecided
_____ Approve
_____ Strongly approve
The person selects their response by checking the appropriate option, which
is then scored 1, 2, 3, 4, or 5, respectively. Likert scales include multiple questions
about the same attitudinal object, and then the responses are summed (or averaged) to produce an overall score. The strength of a person’s attitude is reflected
in the score, ranging from strongly negative = 1 to strongly positive = 5, with all
options in between. The 5-point scale described here was originally developed by
Likert, but subsequent work has resulted in many different Likert-type scales—for
Woo00037_Ch06.indd 190
05-01-2023 18:57:49
CHAPTER 6: Attitudes and Behavior
191
example, scores that range from 1 to 7, scores ranging from –3 to +3, and even
scores that omit the “undecided” option to produce 4-point or 6-point scales (see
Figure 6-4).
FIGURE 6-4
Likert and LikertType scales are
frequently used to
measure attitudes
© Yeexin Richelle/
Shutterstock.com
Another explicit measure used in the study of attitudes is the feeling thermometer. The feeling thermometer is a tool for measuring attitudes that draws
its inspiration from a temperature gauge. The feeling thermometer asks respondents to rate their attitudes toward an object using a scale ranging from 0 to 100.
A score of 0 = very cold, and a score of 100 = very warm. Researchers typically
interpret scores below 50 as unfavorable and those above 50 as favorable. Scores
of 50 are considered neutral, or “no feeling at all.” The feeling thermometer was
first used as a tool in the American National Election Study (ANES), but it has
since been adapted to a range of research areas (Nelson, 2008). To illustrate,
the ANES has consistently used feeling-thermometer measures to gauge public
opinion about a large number of social groups across the United States. Using
data from 2012, the Republican Party scored 47 on the feeling thermometer,
whereas the Democratic Party scored 55. Of the various groups included in the
measurements, the most favorable were the military (score of 83) and workingclass people (score of 84); the least favorable were atheists (39) and the Tea Party
(46; Liu & Wang, 2015).
A key consideration in the development of an attitudinal scale is its degree of
reliability (DeVellis, 2011). As we discussed in Chapter 2, reliability means consistency of measurement. If we measure the same attitude in a person at two very
close points in time, the two measures should be nearly identical. Let’s use a tape
measure as a metaphor for reliability. A highly reliable tape measure—for example, a digital tape measure that uses an infrared beam to detect distance—would
provide high-precision estimates of length. Now consider a tape measure made
of elastic that stretches when pulled. Both the elastic measure and the infrared
measure provide an estimate of distance, but one is much more reliable. In social
psychology, establishing the reliability of a scale is difficult because individuals
Woo00037_Ch06.indd 191
Reliability: means
consistency of
measurement.
05-01-2023 18:57:50
192
Validity: refers to
the accuracy of the
measure.
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
(unlike beams of steel) are prone to change over time. But social psychologists have
developed techniques for determining the reliability of a scale, and research on
attitudes places an especially high value on establishing highly reliable measures.
Explicit measures are by far the most commonly used method for measuring
attitudes. But despite the intuitive appeal of asking individuals to report their attitudes, there are some drawbacks, including lying, carelessness in responses, and
acquiescence (saying “yes” just to appear cooperative). In addition, some attitudes
may not be readily accessible to introspection, and the person may not be consciously aware of their implicit or unconscious associations. These limitations call
into question the validity of the measure. Validity refers to the accuracy of the
measure. One of the primary drawbacks of self-report measures is the tendency
for individuals to respond in socially desirable ways. Social desirability refers to
the tendency to answer questions in ways that will garner approval from others.
Fortunately, techniques have been developed to circumvent social desirability in
responses to attitudinal questions, but it’s important to keep in mind that the limitations of self-report measures are ever-present (Furr, 2011).
GOAL-DIRECTED BEHAVIOR
An alternative approach to measuring attitudes involves inferring them from a person’s actions. Rather than asking a person to self-report their attitude about a topic,
the goal-directed approach focuses instead on a person’s actions and infers the underlying attitude based on their pattern of behavior. Consider the case of a person’s
attitude toward physical fitness. In the approach using goal-directed behavior, we
could ask a person to report whether they had engaged in different behaviors in the
past month—for example, riding a bike to school, walking to the grocery store instead of driving a car, running 3 miles, participating in a spin class, hiking for more
than an hour, going to a gym, and so on. Some of these behaviors require more effort
than others. Riding a bike to school is relatively easy, and it’s likely that many university students will report doing this. By contrast, running 3 miles is much more
difficult, and far fewer people will report doing it. With a large number of responses,
it’s possible to rank the items from easy to difficult. Using these ranked behaviors,
we can then look at each person, and their attitude toward exercise can be inferred
from the most difficult action they report (Kaiser et al., 2010) (See Figure 6-5).
FIGURE 6-5
Goal-Directed Behavior involves inferring a person’s
attitudes from
their actions
Woo00037_Ch06.indd 192
Attitude about physical fitness
Unfavorable
Favorable
05-01-2023 18:57:50
CHAPTER 6: Attitudes and Behavior
193
IMPLICIT ASSOCIATION TEST
Unlike the traditional explicit measurement approach to attitudes, which requires a response to a direct question, implicit measures use decision-making
tasks that allow for inferences about a person’s attitude (Greenwald & Banaji,
2017). Examples of decision-making tasks include classifying words or pictures into different types (e.g., classifying a word as “good” or “bad”) and deciding whether two images are the same or different. One of the most widely
used implicit measures is the Implicit Association Test (IAT; Greenwald
et al., 1998).
The IAT is a classification task that can measure the strength of the cognitive association between pairs of concepts—for example, the stereotypic association between certain genders and careers, such as men and science or women
and nursing. To guide our discussion of the IAT, let’s examine your implicit
attitude toward body weight. Figure 6-6 shows the concept pairs used in an
IAT that is available online through Project Implicit. Before proceeding with
this chapter, we suggest clicking the link below and completing the weight IAT
at Project Implicit:
https://implicit.harvard.edu
After clicking the link and entering the Project Implicit website, select “Social
Attitudes,” review the informed consent and select “I wish to proceed,” and then
select “Weight IAT.” Note that there are many other IATs available, designed to
measure different implicit attitudes, including a Race IAT, a Gender-Science IAT,
and a Sexuality IAT, among others. Before going forward, it is helpful to first complete the online IAT and measure your own implicit attitudes about weight. It takes
about 10 minutes.
In the IAT you just completed, images were used to represent “fat people” and
“thin people.” The second set of categories included “good” and “bad,” represented
by different positive or negative words. For example, delightful or pleasing versus
ugly or rotten. Your task was to classify the words or images into the correct categories as quickly as possible. One word or picture was presented at a time, and the
categories were combined. For example, hurtful: Is this a good word or a bad word?
(The answer is “bad.”) Then the categories were combined: “good or thin” versus
“bad or fat.” In these trials, your task was to classify the word or image into one of
these two combined categories. For example, the word terrific belongs to the “good
or thin” category, whereas hurtful belongs to “fat or bad.” The categories were then
reversed, so your next set of trials used “good or fat” versus “bad or thin.” In this
instance, delightful would belong to “good or fat.”
The IAT is scored using the speed and accuracy with which you make the classifications. Paired categories that have strong positive associations in your cognitive network are easier to classify than paired categories that are not strongly
associated. In our previous example, individuals are generally faster at responding
to “good or thin” versus “bad or fat” pairings than they are at responding to “good
Woo00037_Ch06.indd 193
05-01-2023 18:57:50
194
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
Implicit association test
Next, you will use the ‘E’ and ‘I’ computer keys to categorize items into groups as fast as you can.
These are the four groups and the items that belong to each:
Category
Items
Good
Delightful, Friendship, Magnificent, Pleasing, Cherish, Fabulous, Excitement, Terrific
Implicit attitudes
toward body
weight.
Bad
Ugly, Hurtful, Poison, Rotten, Disaster, Bothersome, Awful, Failure
Source: Project Implicit, available online
at https://implicit.
harvard.edu/implicit/.
Click “Project Implicit
Social Attitudes.” Click
“I wish to proceed.”
Click “Weight IAT.” Click
“Continue.”
Thin people
FIGURE 6-6
Fat people
There are seven parts. The instructions change for each part. Pay attention!
Continue
or fat” versus “bad or thin.” For interpretation, this means that individuals generally have a negative implicit attitude toward heavier body weight. The IAT uses
milliseconds as the measurement unit, then calculates the relative difference between the two sets of pairings. The resulting score represents the preference, and it
generally ranges from 0 to 1 (although IAT scores can often exceed 1). In the case
of implicit attitudes toward body weight, a typical score might be +0.80, meaning
a strong positive attitude toward thinness (and conversely, a strong negative attitude toward fatness). The IAT has become a widely used implicit measure in social
psychology, and a readable overview of this research can be found in Banaji and
Greenwald (2013).
SOCIAL NEUROSCIENCE
An emerging area in the study of attitudes involves monitoring a person’s physiological and neurological responses to an attitude object (cf. Corlett & Marrouch,
2019). The underlying assumption is that individuals have different physiological and neurological responses to stimuli that they like (or agree with) than to
stimuli that they don’t like (or don’t agree with). For example, their face muscles
might become more tense, their pupils might dilate, or different areas of their
brain may become more active. That being the case, we can use instruments to
monitor these responses and thereby measure a person’s attitudes in ways that
don’t require self-reports or even the completion of a specific task—we just present the attitude object and then observe the physiological or neurological response. Various physiological reactions have been used in the study of attitudes,
such as galvanic skin response (GSR; which measures sweating), muscle tension
in the face, heart rate, pupil dilation, and even salivation (Cacioppo et al., 1993;
Ito & Cacioppo, 2007).
Woo00037_Ch06.indd 194
05-01-2023 18:57:51
CHAPTER 6: Attitudes and Behavior
195
FIGURE 6-7
Social Neuroscience involves
monitoring a person’s physiological
and neurological
responses to an
attitude object
© clicksdemexico/Shutterstock.com
Studies using neuroimaging techniques have been particularly useful in this line
of work. Research in this area has shown that positive and negative evaluations are
processed in different areas of the brain, and there is even some evidence that the brain
processes positive information and negative information independently (Cacioppo
et al., 1997). Importantly, research in neuroscience has shown that memories and cognitive associations are not localized, meaning that there isn’t a specific spot in your
brain where your attitudes are stored, much less a specific attitude. That being the case,
it’s not possible (at least with current technology) to simply scan a person’s brain and
identify their attitudes toward different objects. However, with the use of neuroimaging techniques such as functional magnetic resonance imaging (fMRI), researchers
can present a person with an attitude object and then monitor the activation of the
brain areas that are associated with positive or negative evaluations. A person’s attitudes can then be inferred by the change in the activation levels of these regions in response to a specific attitude object—for example, seeing a picture of a political figure,
hearing a popular song, or seeing a specific product (Cunningham & Luttrell, 2015).
With the use of these social neuroscience approaches, it’s possible to measure
attitudes without relying on self-report. But can we perhaps go a step further and
cause specific attitudes to develop? Drawing on research that has mapped regions of
the brain that are associated with positive or negative evaluations, researchers have
begun testing strategies for activating (or inhibiting) these areas and then measuring
the impact on different types of attitudes (Marini et al., 2018). Because we know specific regions of the brain are associated with positive evaluations, research has shown
that it’s possible to induce a positive attitude by activating the area while presenting
a specific attitude object. Using techniques such as noninvasive brain stimulation,
researchers can use electrical current and magnetic pulses to activate or inhibit a specific region of the brain, then examine the impact on subsequent thoughts, feelings,
or behaviors. For example, Sellaro et al. (2015) used noninvasive brain stimulation
to either activate or inhibit the medial prefrontal cortex region. Earlier research has
shown that this region of the brain is less active among individuals who show higher
levels of prejudice (i.e., more negative attitudes toward a specific group). Following
Woo00037_Ch06.indd 195
05-01-2023 18:57:51
196
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
the noninvasive brain stimulation, participants completed an IAT measuring the degree of preference for their in-group (the study was conducted with Dutch students)
versus an out-group (Moroccans). Results showed less prejudice toward Moroccans
among participants with activated medial prefrontal regions compared to participants
who received inhibition of the region, and control participants (Sellaro et al., 2015).
Social neuroscience is a relatively new area within social psychology but one
that offers considerable promise for advancing our understanding of attitudes.
Origins of Attitudes
Up to this point in the chapter, we’ve defined attitudes and discussed some of the
strategies developed to measure them. But where do attitudes come from?
STOP AND THINK: What is your attitude toward dogs? Do you have
a favorable or unfavorable attitude? And how did you come to hold this
attitude?
For the most part, attitudes are learned. That is, our attitudes are a result of
our past experiences. For example, a person who is bitten by a dog as a child may
have negative attitudes toward dogs as an adult. Or alternatively, perhaps you had
a beloved family dog that comforted you during difficult emotional times in your
childhood. In this next section, we briefly summarize the research findings regarding attitude formation. As we’ll see, the process of attitude formation can be quite
complex, and even two people with an identical experience can develop very different attitudes. In addition, there is a close connection between the formation of
an attitude and attitude change. The topics of attitude change and persuasion are
addressed in more detail in Chapter 13 (Persuasion and Compliance).
MERE EXPOSURE
One way in which attitudes can be formed is through mere exposure to an attitude
object over time. For example, think about hearing a new song for the first time. At
first, you’re probably somewhat ambivalent—perhaps you like the genre of music
or even the artist, but you’re not sure about this specific song. But you might find
that over time, the song grows on you, and after a while, you find yourself singing
the words and tapping to the rhythm.
The mere-exposure effect refers to the tendency to develop a more favorable
attitude toward objects that are familiar, even in the absence of a specific positive
experience (Zajonc, 2001). This effect is especially strong for the first few exposures, but favorability continues to increase with continued exposure. Studies in
this area typically use unfamiliar images, such as polygons or symbols from an
unfamiliar language. In these studies, some images are presented more frequently
than others, and then participants rate the degree to which they like each image
Woo00037_Ch06.indd 196
05-01-2023 18:57:51
CHAPTER 6: Attitudes and Behavior
197
or word. Consistently, images that are presented more frequently are rated more
favorably (Bornstein & Craver-Lemley, 2017). There is also evidence showing that
the mere-exposure effect is stronger for images presented subliminally (that is, so
fast that they are difficult to detect consciously; Bornstein & D’Agostino, 1992).
In a classic series of studies by Zajonc (1965), participants were asked to rate unfamiliar words—for example, jandara or lokanta—on a scale of 1 = bad to 5 = good.
They were told, “This is an experiment concerning the effectiveness of repetition
in learning to pronounce strange words correctly.” Some of the words were shown
once, whereas others were shown twice, 5 times, or 10 times. Each participant saw a
different combination of words, some with more frequency than others. The participants were instructed to speak each of the words when they were shown, and after
completing all the trials, they rated each word using a 5-point scale from 1 (bad) to
5 (good). The results are shown in Figure 6-8. The results showed that consistently,
those words that were presented more frequently were more often rated as “good.”
Iktitaf
Low frequency
High frequency
Afworbu
Saricik
Biwojni
Nansoma
Kadirga
Enanwal
Dilikli
Zabulon
FIGURE 6-8
The mereexposure effect.
Lokanta
Jandara
Civadra
1
2
3
4
Rated “goodness” of meaning
Woo00037_Ch06.indd 197
5
Notes: Results from
a study showing that
unfamiliar words that
are seen more frequently are rated more
positively than words
that occur with low
frequency.
Source: Redrawn from
Zajonc (1968).
05-01-2023 18:57:52
198
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
As an easy extension of the mere-exposure effect, think about which letter of the English alphabet is your favorite. Stop reading for a moment and
choose a letter. Now read on. For this author, my favorite letter is s. It has
many interesting and positive features, but not surprisingly, it also occurs in
both my first name (Wesley) and my last name (Schultz). In fact, it’s common
for people to prefer letters from their names and numerals from their birth
month or date.
The mere-exposure effect has implications across a number of applied areas,
from dating and interpersonal attraction to health and the marketing of consumer
products. Here are just a couple of examples:
Woo00037_Ch06.indd 198
•
Personal attraction. Propinquity refers to the tendency for individuals
to be more attracted to people they encounter more often—for example, attraction to a work colleague you see every day or to a person living nearby in your apartment complex. Festinger et al. (1950) showed
that residents in an apartment complex who lived near the stairs tended
to have more friends on other floors than residents who lived farther
from the stairs. This is because residents who lived closer to the stairs
were closer in proximity to residents on other floors than residents
who lived farther from the stairs, and they were therefore more likely
to see these residents. We will return to this study in Chapter 9 – Liking
and Loving.
•
Getting children to eat more vegetables. Ahern et al. (2014) presented
children with exposure to novel vegetables in their nursery schools and
then measured their intake of similar vegetables 6 months later. The
authors concluded, “This study confirms previous observations that
repeated exposure increases intake of a novel vegetable in young children” (p. 154).
•
Attitudes about political candidates. During election season, it’s common
to see thousands of posters, advertisements, and lawn signs supporting a
political candidate or ballot measure. But are these signs effective? Consider the case of lawn signs. Clearly, the person putting the sign in front
of their house already has a favorable attitude toward the candidate, but
what effect do these signs have on other voters? Research suggests that
the presence of these signs can influence vote share for the candidate.
To illustrate, consider a series of studies by Green et al. (2016) in which
lawn signs were placed at higher rates in randomly selected voting districts, followed by tracking of the votes across all the districts. Because of
the random assignment of areas with more signs, there should have been
no difference in the voting rates of the chosen districts, but the results
showed a clear increase in vote share for those districts selected for the
increased number of signs.
05-01-2023 18:57:52
CHAPTER 6: Attitudes and Behavior
199
FIGURE 6-9
Social psychologists measure the
influence of lawn
signs on voter
attitudes
© Sharkshock/Shutterstock.com
CLASSICAL CONDITIONING
Classical conditioning is the type of learning originally investigated by Ivan Pavlov, who paired meat powder given to dogs with the sound of a bell. Over a series
of trials, he found that the bell alone would cause the dogs to salivate. In classical conditioning, an unconditioned stimulus (UCS) can automatically produce
an unconditioned response (UCR); in the case of Pavlov’s dogs, the meat powder
(UCS) caused the dogs to salivate (UCR). No training was needed for the salivation to occur. But then he used a new stimulus, the bell, which did not automatically elicit salivation, and he presented this just before the UCS on several trials.
To test whether learning had occurred, the UCS was then omitted. In studies such
as these, if the response is given to the other stimulus alone, we say that the response has been conditioned to that stimulus, which is termed the conditioned
stimulus (CS).
Applying this type of learning to attitude formation, we might consider a parent who scolds their son in a firm tone. The firm tone of voice automatically produces negative, unhappy feelings in the child. If the scolding is applied every time
the child reaches for a valuable vase, the vase will soon become a CS that will produce negative feelings (an attitude). Of course, positive attitudes (for instance, toward Grandma) can be produced in the same way by frequently pairing the object
(Grandma) with a UCS, such as a tasty treat or a hug, that makes the child feel good
(Oskamp & Schultz, 2005).
Woo00037_Ch06.indd 199
05-01-2023 18:57:52
200
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
FIGURE 6-10
Attitudes can
be learned
through classical
conditioning
Note: Classical conditioning is the type
of learning originally
investigated by Ivan
Pavlov, who paired
meat powder given to
dogs with the sound of
a bell.
© Drp8/Shutterstock
.com
Researchers have used the classical conditioning paradigm to investigate a
range of topics (Vogel & Wänke, 2016). As an example, Olson and Fazio (2001)
presented participants with novel stimuli (Pokémon characters) paired with either
positive words (“excellent,” “awesome”) and images (puppies, a hot fudge sundae)
or negative words (“terrible,” “awful”) or images (a cockroach). Note that at the
time of this study, Pokémon characters had just been introduced in the United
States, and very few people were familiar with them. After viewing these pairings
repeatedly on a computer screen, participants rated the Pokémon characters that
had been paired with positively valenced stimuli more favorably than the Pokémon
characters that had been paired with the negative stimuli. That is, by pairing the
CS (Pokémon characters) with a UCS (positive or negative words or images) over
many trials, the CSs (Pokémon characters) were eventually able to elicit the CRs
(positive or negative evaluation). In a parallel study involving the classical conditioning of implicit attitudes, the researchers found the same effect, but somewhat
stronger, when using the IAT as the attitude measure (Olson & Fazio, 2001).
Importantly, in both of these studies, there was no evidence that the participants were aware of the newly developed associations. This fact is critical because
a major question in performing classical conditioning is whether participants
are aware of the stimulus–response connection that the investigator is trying to
Woo00037_Ch06.indd 200
05-01-2023 18:57:52
CHAPTER 6: Attitudes and Behavior
201
establish. If they are aware, any learning may be of an instrumental sort (trying to
do what will gain the researcher’s approval), or alternatively, it may be of a cognitive, information-processing sort rather than being classical conditioning.
OPERANT CONDITIONING
Operant conditioning is the type of learning investigated by Skinner (1974). It is
called operant because the organism is allowed to operate freely in its environment
instead of being constrained to make one particular response or pair two stimuli
together. In essence, operant conditioning involves rewards and punishments that
are experienced as a consequence of the action. If a behavior is rewarded, it is more
likely to occur in the future. A behavior that is punished is less likely to occur. In
studies of operant conditioning, the researcher does not usually know or care what
the original stimulus was for the organism’s behavior. Instead, the researcher waits
until the organism makes a desired response (say, a rat scratching its ear) and then
immediately reinforces the response by presenting a desirable consequence, such
as a food pellet. Over time, the frequency of the rewarded behavior will increase.
In human attitude formation, the reinforcer is apt to be verbal—either praise or
criticism—or nonverbal signs of approval or disapproval. For instance, a child might
say, “Fat people are gross” and be rewarded by an approving smile from the parent.
As a result, the child will likely make other disparaging weight-based comments in
the future, and they might also form a negative attitude toward heavier body weight.
OBSERVATIONAL LEARNING—IMITATION AND MODELING
Observational learning involves imitation of the behavior of another person who
serves as a model and can occur without any external reinforcement. Parents are
often pleased to find that their children imitate their admirable behavior (e.g., helping to feed the baby) but are also concerned when their children imitate their antisocial acts, such as cursing at a disliked neighbor or flipping off a motorist on the
highway. In many such modeled actions, the behavioral aspect of attitudes begins
to be formed without any explicit instruction or reinforcement by the parent. In
fact, children will often imitate what they see their parents do instead of what they
say (Rushton, 1975). Much research has shown the effectiveness of models in shaping attitudes and behavior—for instance, in the areas of aggression (Bandura, 1977;
Sparks & Sparks, 2002), helping behavior (Berkowitz, 1972), and attitudes toward
learning and science (Fisch, 2002).
As an example, consider the case of attitudes toward smoking cigarettes among
adolescents. In television and (especially) movies, smoking cigarettes is often presented in a favorable light. Well-known and glamorous celebrities can often be seen
smoking in a social context or as a way to cope with a stressful situation. Observational learning would suggest that viewing such models would cultivate a favorable attitude toward smoking cigarettes. In a large correlational study, Sargent et al.
(2001) measured the amount of smoking in 601 popular movies. Among a sample
Woo00037_Ch06.indd 201
05-01-2023 18:57:52
202
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
of 3,766 middle school students (ages 11–14), each student reported their exposure
to various movies. The results showed that students who reported seeing movies
with more tobacco use had more favorable attitudes toward smoking cigarettes,
and in addition, students with greater exposure to movie-related tobacco use were
more likely to report having tried cigarettes themselves. Similar results have been
found for exposure to advertisements for e-cigarettes among adolescents (Singh
et al., 2016). See also the discussion in Chapter 11 (Aggression and Other Anti-Social Behavior) regarding exposure to media and video-game depictions of aggression and violence.
STOP AND THINK: List three of your strongest attitudes. When did you
form this attitude?
IMPRESSIONABLE YEARS
Impressionableyears hypothesis:
young adulthood is an
especially important
period for the
formation of attitudes
Woo00037_Ch06.indd 202
The impressionable-years hypothesis suggests that young adulthood is an especially important period for the formation of attitudes and that after this period, susceptibility drops and remains low for the rest of the person’s life. The classic study in
this area is the Bennington College study (Newcomb, 1943), in which selected incoming female college students were asked about their social and political attitudes
before, during, and after their college years. The students at Bennington College
were raised in politically conservative and wealthy households but found themselves in a very liberal college environment at Bennington. The results from the
study showed that over their 4 years at Bennington College, the students became
progressively more liberal. Even more importantly, when reinterviewed 25 years
after they graduated from Bennington, most remained liberal, despite the influence
of their conservative families (Newcomb, 1967). An even longer-term follow-up of
these students found that this political liberalism acquired while at college continued 50 years after graduation (Alwin et al., 1991). Although providing some initial
insight into the impressionable-years hypothesis, these data are limited by the small
sample size and the focus on a single cohort of students.
More systematic support for the impressionable-years hypothesis can be
found in data from the ANES. The ANES is an ongoing survey of a representative
cross-section of adults living in the United States. Krosnick and Alwin (1989) reported analyses using data from 1958 to 1976, with seven different age groups (“impressionable adults” were in the age category of 18–25). The researchers focused on
five different types of attitudes: liberal–conservative ideology, attitudes toward social groups, attitudes about domestic policy issues, and attitudes toward prominent
politicians. Their results showed that the youngest age group (ages 18–25) had attitudes that were less stable over time than did the other six age groups. There were
no differences in attitude stability outside of the youngest age group, suggesting
that once the “impressionable years” pass, attitudes do not become more resistant
05-01-2023 18:57:52
CHAPTER 6: Attitudes and Behavior
203
to change (Alwin & Krosnick, 1991). Other studies have replicated this effect and
have also extended the findings into other domains (Giuliano & Spilimbergo, 2010).
Let’s consider a final example in which we use the impressionable-years hypothesis to explain preferences for specific products. Think for a moment about the many
products that you consume on a daily basis—for example, coffee or potato chips.
When did these preferences begin, and how stable are they over time? In an analysis
of taste preferences, Sudhir and Tewari (2016) examined the consumer choices of
more than 100,000 individuals from across China over a 20-year period, from 1992
and 2011. They found strong evidence that “millennials”—individuals born during
the end of the 20th century—were more likely to purchase nontraditional Chinese
products, such as coffee, packaged snack foods like potato chips and pretzels, and fast
food such as KFC and McDonald’s. In addition, and in line with the impressionableyears hypothesis, once these food preferences were formed, they persisted over a 20year period. Those individuals born during the “millennial” period came to prefer
these new foods over traditional Chinese foods, and these preferences continued into
later life. By comparison, older generations who were raised on traditional Chinese
foods were less likely to consume the newly introduced Western foods (for example,
coffee) and instead preferred the traditional Chinese diet (for example, tea).
FIGURE 6-11
Do you prefer
coffee or tea?
Note: Research shows
that a person’s preference for coffee or tea,
formed during their
impressionable years as
young adults, is likely
to persist across their
adult lives.
© YaiSirichai/Shutterstock.com
Do Attitudes Predict Behavior?
One of the central issues for attitude researchers is the relationship between attitudes and behaviors. Intuitively, attitudes should be good predictors of behavior. A person’s actions should follow from their attitudes.
Woo00037_Ch06.indd 203
05-01-2023 18:57:53
204
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
However, research on the relationship between attitudes and behavior suggests that the link is not always straightforward. Consider the following
three examples:
Woo00037_Ch06.indd 204
•
University students were asked about their attitudes toward eating a range
of different exotic foods, including snakes, octopus, grasshoppers, and
others. Grasshoppers were rated as the most disliked. Students were then
asked either individually or in a group of 10 others to eat a grasshopper.
The experimenter said, “Before asking you to eat the experimental food, I
want to make it clear that this part of the experiment is voluntary, and no
one has to eat the fried grasshoppers if they don’t want to.” Yet despite the
voluntary nature of the food offering and their strong negative attitudes,
approximately half (50%) ate the grasshopper (Zimbardo et al., 1965). The
study was replicated with samples of soldiers serving in the U.S. Army Reserve in a study of survival foods, with similar results. In these instances,
despite their negative attitudes toward certain foods, participants still engaged in the behavior of eating those foods.
•
A group of homeowners participated in a 3-hour workshop about water and energy conservation, during which they received a free lowflow showerhead. Following the workshop, the participants showed
favorable attitudes toward conservation efforts, and they were more
knowledgeable about ways to conserve water and energy in their
homes than they were at the start. Yet in a 6-week follow-up visit to
the homes, only 8 in 40 (20%) had actually installed the showerhead,
and even fewer had done any of the other recommended home actions (Geller, 1981). In this example, despite the favorable attitudes
toward water and energy conservation, few people acted on their
convictions.
•
In 1934, a researcher (LaPiere) drove across the United States with a
Chinese couple, stopping at hotels and restaurants along the way. Keep
in mind that in 1934, many people in the United States had strong negative attitudes toward Chinese people, and it was commonplace for
restaurants to refuse service based on race and ethnicity. Yet during their
10,000-mile cross-country trip, they stopped at more than 250 hotels
and restaurants, and they were only refused service once. They were rejected from 1 of 66 hotels and served at all 184 restaurants (in 72 of them,
they reported receiving good service). Six months later, LaPiere sent a
questionnaire to each of the restaurants and hotels to ask if they would
“accept members of the Chinese race as guests in your establishment.”
Of the 128 that responded, 92% reported that they would not. Clearly,
this is an instance where the attitudes of restaurant owners didn’t match
their actions.
05-01-2023 18:57:53
CHAPTER 6: Attitudes and Behavior
205
STOP AND THINK: Can you recall a time when your behavior didn’t
match your attitude? (Examples include acting friendly toward someone
you dislike or eating a food that you think is disgusting.) Why do you
think you acted as you did? And did your attitude change afterward?
Although these examples clearly show that individuals can act in ways that are
inconsistent with their attitudes, research has shown that in general, attitudes can
predict behavior, and in some situations, they are strongly predictive (Ajzen et al.,
2019; Glasman & Albarracín, 2006). As we’ll see, research has identified several situational and personal factors that help to determine whether the relationship will
be strong or weak. Variables such as the strength of the attitude, how easily it can be
activated, how salient a particular attitude is in a given situation, and how relevant
the attitude is to the particular behavior all help determine whether people will act
in accordance with their attitudes (Eagly & Chaiken, 1993).
Personality traits are also important. Some people are more likely to match their
actions to their attitudes. Others have a tendency to override their own attitudes in
order to behave “properly” in a given situation. As a result, attitudes do not predict
behavior as well for some people as for others (Snyder & Tanke, 1976). In particular, people who are high on self-monitoring are especially likely to override their
attitudes to behave in accordance with others’ expectations. Before speaking or acting, high self-monitors observe the situation for cues about how they should react.
They then try to meet those situational “demands” rather than behave according
to their own beliefs or sentiments. In contrast, low self-monitors express and act
on their attitudes with great consistency, showing little regard for situational cues
or constraints. For example, a high self-monitor who disagrees with the politics of
a respected dinner guest may keep her thoughts to herself in an effort to be polite
and agreeable, whereas a low self-monitor who disagrees might dispute the speaker
openly, even though doing so might disrupt the social occasion (Snyder, 1987).
RESEARCH ON EXPLICIT MEASURES OF ATTITUDES
In a meta-analysis of the research on the relationship between attitudes and behavior,
Kraus (1995) found that attitudes were strongly and positively related to behavior. Recall that meta-analysis combines the findings from across a large number of studies in
order to synthesize the results and provide a cumulative finding. Kraus identified 119
applicable studies, from which he was able to extract 88 correlation coefficients. Across
these 88 correlation coefficients, the average was r = .38, with a range from –.10 to +.91.
Recall that this is a medium to large effect size. Across the studies, he found that the
relationship between attitudes and behavior was stronger in certain circumstances:
•
Woo00037_Ch06.indd 205
When attitudes were formed through direct experience. Examples include a child who works on a puzzle and then forms a negative attitude
toward the puzzle because of its difficulty. This direct experience can be
05-01-2023 18:57:53
206
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
contrasted with a child who develops a negative attitude toward the puzzle
based on a conversation with a peer who found it difficult.
Woo00037_Ch06.indd 206
•
When attitudes were more accessible (defined as easy to recall). For
example, how quickly does a person respond when asked about their attitude? Faster responses indicate a more accessible attitude, and more accessible attitudes tend to correlate more strongly with behavior.
•
When attitudes were more certain. For example, consider an upcoming election for student class president. Students are first asked about their attitudes
toward the various candidates, and for each, they are also asked, “How certain
are you about this attitude?” Overall, a person’s attitudes tend to be correlated
with their subsequent vote in the election. But for students who report being
highly certain about their attitude, the relationship tends to be even stronger.
•
Among individuals who were low in self-monitoring. As discussed
earlier, individuals who are high in the self-monitoring personality characteristic tend to modify their behavior to match the demands of the context. Given this description, it’s not surprising that people who are high
self-monitors show a weaker correspondence between their personal preferences and behavior—their behavior is more subject to the demands of
the context than to personal convictions.
•
When behaviors were measured with self-report rather than observed.
Explanations for this finding center around the tendency for individuals to recall behaviors in a way that is consistent with their attitudes. No
one wants to be a hypocrite, where they say one thing but do another, and
there is pressure for our attitudes and behaviors to be consistent. This is
especially the case when the topic is private or socially desirable, such as
attitudes toward safe-sex practices, racial attitudes, or attitudes toward a
contentious politician.
•
Among nonstudent populations. One explanation for this finding is that
students tend to have attitudes that are still in flux and developing, and
many of their attitudes may not be based on direct experience. Note that
this explanation is in line with the impressionable-years hypothesis discussed earlier (and it also presumes that students tend to be younger than
the general public).
•
When attitudes and behaviors were measured at “corresponding levels
of specificity.” Specificity here refers to the similarity between the measured
attitude and the behavior. For example, in the study described previously
that gave participants low-flow showerheads, the participants’ attitudes
toward conserving water and energy were measured. Behavior was then
measured as the installation of a low-flow showerhead. This represents a
low level of corresponding specificity, and as we saw, attitudes in this situation were not predictive of behavior. In contrast, had the researcher asked
about attitudes toward low-flow showerheads, it’s likely that these attitudes
would have been more predictive of installation behavior.
05-01-2023 18:57:53
CHAPTER 6: Attitudes and Behavior
207
Taken together, the results show that overall, a person’s attitudes are generally
predictive of their actions (Ajzen et al., 2019). But importantly, just because attitudes and behaviors are correlated, it does not mean that attitudes cause behavior.
It may be that the direction of causality is reversed—first, the person engages in
the behavior, and then the individual forms an attitude to match. Or alternatively,
both attitudes and behavior could be caused by a third variable, such as a genetic
predisposition to prefer certain types of experiences.
IMPLICIT ATTITUDES AND BEHAVIOR
Earlier in this chapter, we introduced the idea of implicit attitudes and used the IAT
as an example of the measurement of such attitudes. Implicit attitudes are positive
or negative evaluations of an attitude object that occur without conscious awareness (Greenwald & Banaji, 2017). This is contrasted with the more traditional explicit attitude, which is typically measured by asking for self-reported responses.
A meta-analysis of studies that used the IAT identified 122 research papers that reported 526 correlations between implicit attitudes and behavior (Greenwald et al.,
2009). The results showed that implicit attitudes were significantly and positively
correlated with behavior (r = .27). The correlation is slightly smaller than that reported for the relationship between explicit attitudes and behavior (r = .38), as
described previously. The results also showed that implicit measures were better
predictors of certain types of behaviors, especially those that were related to socially sensitive topics, such as race, gender, and sexual orientation. The explanation
for the stronger results with the socially sensitive topics is that individuals tend
to censor their responses when asked to self-report, but implicit attitudes are less
controllable.
To illustrate the predictive power of implicit attitudes, let’s consider a study of
newlywed couples who completed both implicit and explicit measures about their
attitudes toward their relationship and their partner (McNulty et al., 2013). The researchers then contacted each individual every 6 months for the next 4 years, with
the same implicit and explicit measures, along with measures of their relationship
satisfaction, specific problems in their marriage, and whether or not they were still
together. The researchers found that there was no correlation between implicit and
explicit attitudes and noted that this “suggests that spouses were unaware of their
automatic attitudes” (p. 1119). The results also showed that over the 4-year period,
marital satisfaction declined. But interestingly, this decline in marital satisfaction
was stronger for those individuals who initially had more negative implicit attitudes toward their partners. Individuals who had a more positive implicit attitude
toward their partners at the time of marriage had happier marriages 4 years later.
For explicit attitudes, this was not the case, and individuals with more favorable
explicit attitudes at the time of marriage were more dissatisfied 4 years later. This
study illustrates two basic points highlighted in this chapter. First, both implicit
and explicit attitudes can predict future attitudes and behaviors—in this case, marital satisfaction. And second, implicit and explicit attitudes operate differently. We
will examine implicit and explicit attitudes as they relate to prejudice in Chapter 8.
Woo00037_Ch06.indd 207
05-01-2023 18:57:53
208
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
EXPECTANCY-VALUE THEORY
Throughout this chapter, we’ve discussed attitudes as evaluations that range from
positive to negative. Expectancy-value theory builds on this framework by incorporating a person’s perceptions about the likelihood that engaging in the target behavior will result in a positive outcome (Ajzen & Fishbein, 2008). An expectancy is
a person’s belief about the future. It refers to the perceived likelihood of a specific
outcome. For example, consider a college course in which the professor grades using
multiple-choice exams. Stephen is a student in the class, and he believes that he is not
good at taking standardized tests and expects that he will do poorly on the first exam.
The value aspect of the expectancy-value theory refers to the perceived importance of an outcome (Eccles & Wigfield, 1995; Wigfield et al., 2009). Continuing
with our example of Stephen, although he believes that he has poor test-taking
skills, he also places low importance on doing well in his courses. As he sees it, “Cs
get degrees.” The “value” part of the expectancy-value theory reflects a person’s attitude toward the outcome, and the combination of value and expectancy influences
motivation and, ultimately, behavior. That is, individuals are more likely to engage
in behaviors that they believe are likely to lead to a desired outcome. In Stephen’s
case, he is likely to choose courses that don’t grade using multiple-choice exams,
and when he is in these types of courses, he likely will not engage in the actions
needed to get a good grade. The model has wide-ranging implications and has been
especially useful in the areas of academic achievement, school performance, and
work (Wigfield & Eccles, 2000). We will examine how values and expectancies relate to academic achievement in Chapter 14 (Motivation and Achievement).
THE THEORY OF PLANNED BEHAVIOR
Theory of planned
behavior: According
to the theory of
planned behavior, the
best predictor of a
person’s behavior is
their intention to act.
Woo00037_Ch06.indd 208
The theory of planned behavior is one of the most widely cited and studied theories of attitudes and behavior. As its name implies, the theory proposes that individuals generally act in ways that are planned and thought-out.
According to the theory, the best predictor of a person’s behavior is their intention to act (Ajzen, 2005; Fishbein & Ajzen, 2009). For instance, whether or not a
person uses a condom during sexual intercourse is best predicted by their intention
to do so. These intentions in turn are determined by a person’s attitudes, normative beliefs, and perceived control. Attitudes reflect a person’s positive or negative
evaluation of performing the behavior based on the expected consequences. For
example, a woman may evaluate using condoms positively because of their protective benefits, or she may have negative attitudes toward condoms because of
the potential embarrassment of asking her partner to use a condom. Subjective
norms are a person’s perceptions of the social pressures to perform the behavior.
For instance, what do other people think about using condoms? Finally, the third
predictor of intentions is perceived control, a person’s beliefs about their ability to
perform the behavior. For example, does the person know where to buy condoms
or how to use them? A graphical representation of the theory of planned behavior
is shown in Figure 6-12.
05-01-2023 18:57:53
CHAPTER 6: Attitudes and Behavior
209
Behavioral
attitude
Subjective
norms
Intention
Behavior
FIGURE 6-12
Perceived
behavioral
control
The theory of
planned behavior.
Source: Ajzen (1991)
Putting Research into Action. A large body of research has been conducted
using the theory of planned behavior, and in general, the results provide strong
evidence that it can be used to predict behavior. It has been used effectively to predict
behaviors ranging from engaging in premarital intercourse to using birth control
pills, attending church regularly, and buying many types of consumer products. In
these studies, the average correlation coefficients for predicting intentions using
attitudes was as high as +.8, and the correlation between intentions and behaviors
ranged from +.4 to as high as +.9 (Ajzen, 2005). Let’s look at a specific example.
In the 1990s, HIV infection rates were rising worldwide, and there was no
known treatment or vaccine. In an effort to slow the spread of the disease, researchers conducted a large number of studies focused on understanding and changing the
behaviors that put individuals at risk. HIV infections occur through the exchange
of bodily fluids with an infected individual, particularly with regard to unprotected
sexual activity and sharing needles during injectable drug use. In a large-scale
application of social psychological theory, the AIDS Community Demonstration
Project (ACDP) implemented a community intervention program that aimed to
reduce risky behaviors among individuals who were in high-risk groups for contracting HIV (e.g., injectable drug users and their female sex partners, female prostitutes, and men who have sex with men). The project was implemented in five
major metropolitan areas across the United States (Dallas, Denver, Long Beach,
New York City, and Seattle), reaching tens of thousands of individuals. We’ll focus
here on the results for female sex workers (Jamner et al., 1998).
Interviews were conducted with 634 women who were recruited individually
on the streets in neighborhoods having a high prevalence of drug use and/or prostitution. Among the sample of female sex workers, 25% were married, and 51%
reported having a “main partner.” Only 6% of the women reported using a condom
every time with their husbands or main partners, and only 27% reported using
condoms every time with their paying partners. Items in the questionnaire were
designed to measure each aspect of the theory of planned behavior. Intentions to
use a condom for vaginal sex with main and paying partners were measured using
a 7-point scale. A sample item is, “How likely do you think it is that from now
Woo00037_Ch06.indd 209
05-01-2023 18:57:53
210
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
on, you will use a condom every time you have vaginal sex with your main partner?” Responses were on a scale from –3 (extremely sure I won’t) to +3 (extremely
sure I will). A similar 7-point scale was used to assess attitudes toward condom
use—good versus bad, wise versus foolish, pleasant versus unpleasant, like versus
dislike, easy versus difficult? Subjective norms were measured with questions like,
“Do most of the people who are important to you think that you should or should
not use a condom for vaginal sex with your paying partners?” Finally, perceived
behavioral control was assessed with questions like, “If you wanted to use a condom
every time you have vaginal sex with your paying partner, how sure are you that
you could?”
Analyses tested the ability of attitudes, subjective norms, and perceived control
to predict respondents’ intentions to use condoms with their paying partners. Results showed that 47% of the variability in intentions could be explained using the
three predictor variables. The relationship between attitudes and intentions to use
condoms was r = .37 (p < .001), the relationship between subjective norms and intentions to use condoms was r = .01 (not significant), and the relationship between
perceived control and intentions to use condoms was r = .40 (p < .001). The authors
concluded that “the results from this study provide support for the utility of the
Theory of Planned Behavior as a framework for investigating the factors that influence female sex workers to form intentions to ensure that their main and paying
male partners use condoms” (Jamner et al., 1998, p. 200).
Behaviors Predict Attitudes
In the preceding section, we saw that attitudes can predict behavior. A final question that we need to address in this chapter is the reverse: Do behaviors predict
attitudes? Perhaps people act first, then form or change their attitudes after the
behavior occurs. For example, suppose that a friend of yours (Jordan) is attracted
to another friend of yours (Alex) but decides not to ask them out on a date. Jordan
has a favorable attitude toward dating Alex but has not acted in a way that is consistent with the attitude. Would Jordan’s attitude then change to reflect this decision?
Jordan might think, “Perhaps Alex isn’t such a great person after all.”
COGNITIVE DISSONANCE THEORY
Cognitive
dissonance:
Inconsistency between
our thoughts and
behaviors leads to
a state of cognitive
dissonance—an
uncomfortable state
of tension that we are
motivated to reduce.
Woo00037_Ch06.indd 210
Cognitive dissonance theory proposes that people are motivated to think and
act in a consistent manner. That is, our attitudes and behaviors should be similar.
Inconsistency between our thoughts and behaviors leads to a state of cognitive
dissonance—an uncomfortable state of tension that people are motivated to reduce. For example, the cognition “I would like to go out with Alex” and the behavior “I didn’t ask Alex out” are inconsistent and might result in dissonance. It is
important to note that dissonance results from a psychological inconsistency (i.e.,
a perceived inconsistency) and not necessarily a logical inconsistency (Festinger,
1957, 1964).
05-01-2023 18:57:53
CHAPTER 6: Attitudes and Behavior
211
Once dissonance is produced, the person is motivated to get rid of it. There are
a number of ways to reduce dissonance. One way is to change the behavior in the
future to be consistent with the dissonant cognition. For example, Jordan can text
Alex to meet for coffee before class. A second way to reduce dissonance is to change
the thought (e.g., the attitude) to be consistent with the behavior—for instance, “I
don’t want to go out with Alex anyway” or “Alex really is not that attractive.” A third
reduction strategy is to create new cognitions that justify the inconsistency, such as,
“I’m a very choosy person,” or “First, Alex should ask me out.”
Let’s now look at examples of each of these three dissonance-reducing strategies:
1. Change the behavior to be consistent with the attitude. Let’s continue with the
theme introduced earlier and ask whether we can use cognitive dissonance theory
to promote safer sex practices. In these studies, the researcher examined the effect
of hypocrisy—one form of cognitive dissonance—on condom use among college
students (Stone et al., 1994). Participants in the study were sexually active, heterosexual undergraduates.
Cognitive dissonance was induced by obtaining a public statement regarding
favorable attitudes toward condom use and then making participants aware that
their behaviors were inconsistent with this statement (Aronson et al., 1991). For the
commitment manipulation, participants were asked to develop a persuasive speech
about sexually transmitted infections and the importance of protected-sex practices
and then deliver this speech in front of a video camera. They were told that the
researcher was trying to find the best communicator to get the message about the
importance of protected-sex practices out to high school students and that their
video would be used in a sex education program. Participants who were not in the
commitment condition prepared the speech but were not asked to create a video
about the topic. A second independent variable involved making the participants
aware that they did not always engage in the target behavior—that is, that their own
actions were inconsistent with the speech and/or video. Participants in the “mindful” condition were asked to make a list of the circumstances surrounding their own
failure to use condoms in the past. The researcher said that these were needed to
“help high school students deal more effectively with these situations” (Stone et al.,
1994, p. 119). Participants who were not in the mindful condition were merely given
information about sexual infections and the importance of protected-sex practices.
Thus, the study had four conditions: commitment plus mindful (the dissonance
condition), commitment only, mindful only, and a control group. Participants in
the dissonance condition had the following cognitions: “Using condoms during sex
is important—I just told high school students that I think so—but I have not always
used a condom myself.”
Several dependent measures were obtained—we will focus on two. The first
was a measure of condom buying. At the end of the study, participants were given
a participation credit slip, handed four $1 bills, and asked to fill out a receipt. The
experimenter then said:
Before you start that, let me tell you that the … Health Center sent
over some condoms and pamphlets … when they heard about our
Woo00037_Ch06.indd 211
05-01-2023 18:57:53
212
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
prevention program. They wanted us to give our subjects an opportunity to buy condoms for the same price they are sold at the health
center—10 cents—and this way you don’t have to go across campus
and stand in a long line. I need to go next door and prepare for the
next subject, so go ahead and finish this receipt; you can leave it here
on the table. And if you want to buy some condoms or take some
pamphlets, just help yourself to anything on the desk; that dish has
some spare coins so you can make change. OK? Thanks again for
coming in today. (Stone et al., 1994, p. 119)
FIGURE 6-13
Using cognitive
dissonance to
increase safer sex
practices
Source: Stone, J.,
Aronson, E., Crain, A.
L., Winslow, M. P., &
Fried, C. B. (1994).
Inducing hypocrisy as a
means of encouraging
young adults to use
condoms. Personality
and Social Psychology
Bulletin, 20(1), 116–
128. doi:10.1177/
0146167294201012
Woo00037_Ch06.indd 212
Percentage of subjects who purchased condoms
The experimenter then left the room. After the participant left, the researcher
returned and determined whether the person had purchased any condoms or taken
any of the pamphlets. A second dependent measure was self-reported condom use
after participating in the study. Ninety days after the experiment, participants were
interviewed over the telephone and asked about their sexual behavior since the
study.
The results from the experiment showed that participants in the dissonance
condition changed their behavior to be consistent with their publicly stated attitude. In the hypocrisy condition, 85% purchased a condom, compared with 33%
in the commitment-only condition, 50% in the mindful-only condition, and 44%
in the control condition (see Figure 6-13). Participants in the hypocrisy condition
also purchased more condoms (M = 4.95) than those in the commitment-only
(M = 3.50), mindful-only (M = 2.40), or control condition (M = 3.50).
Follow-up data were obtained from 64 of the original 72 participants. Of these,
52 reported having sex at least once in the 90 days since the study. Three of the
participants reported having extreme levels of sexual activity—sexual intercourse
90 or more times since the study—and were excluded from the analysis. When the
remaining 49 were asked about condom use, 92% of the participants from the dissonance condition reported using a condom, a figure that was significantly higher
than that for participants in the commitment-only condition (55%), mindful-only
condition (71%), and control condition (75%).
100
90
80
70
Hypocrisy
Commitment-only
Mindful-only
Information-only
60
50
40
30
20
10
0
05-01-2023 18:57:54
CHAPTER 6: Attitudes and Behavior
213
These results suggest that inducing cognitive dissonance can be an effective
technique for increasing condom use among sexually active college students. Participants in the study who performed the behavior of publicly stating that using condoms was important and were made mindful of the fact that they did not always do
so were more likely to purchase a condom, purchased more condoms, and reported
more condom use 90 days later than did participants in the commitment-only,
mindful-only, or control conditions.
2. Change the attitude to be consistent with the behavior. Cognitive dissonance
theory was originally developed by Leon Festinger, and the basic principles were
clearly demonstrated in his classic “peg board” study. In the famous Festinger and
Carlsmith (1959) study, research participants arrived at a study labeled as “measures of performance.” During the study, participants were asked to complete a
series of tasks, including putting round spools on a stick and turning square pegs
one-quarter-turn clockwise. Then, after all pegs had been turned a full revolution,
there was a second round in which they turned the pegs counterclockwise. These
tasks lasted for an hour, and according to Festinger, they were some of the most
boring and meaningless tasks he could devise. After completing the tasks, the participants were asked about their attitude—specifically, “Were the tasks interesting
and enjoyable?”—and asked to respond using a scale from –5 to +5. Not surprisingly, the tasks were generally rated negatively.
After completing the tasks, participants were randomly assigned to one of
three experimental conditions. In the first condition, participants were induced to
tell another person that the study was, in fact, “fun and interesting.” This was accomplished by using a cover story in which each participant was told that waiting
in the room next door was the next participant, but unfortunately, the assistant
who was supposed to help with the study hadn’t arrived, and the researcher needed
some assistance. According to the researcher, some of the participants completed
the tasks “cold,” meaning that they weren’t given any prior information about the
tasks. Others, like the one in the waiting room, were supposed to complete the tasks
with positive expectations that the tasks were, in fact, fun and interesting. The participants were then asked if they would be willing to help the researcher and serve
as the assistant for the next participant. Additionally, for their time, the researcher
offered $20 (the equivalent of about $180 today). All agreed to do so, and they proceeded to tell the next participant how fun and interesting the tasks were (clearly
a lie—the tasks definitely weren’t fun and interesting). In the second experimental
condition, participants were offered $1 to tell the lie, and again, all agreed to do so.
Finally, the third condition was a control in which they were not asked to tell the lie.
Before we summarize the results, imagine that you are a participant in the study.
You just completed an hour-long series of very boring tasks, but you’ve now told
someone else that it was fun and interesting. Would this action of telling someone
else that it was fun and interesting affect your attitude? And would it matter that
you were paid $180 to do it? The results are summarized in Figure 6-14. As shown,
participants in the control condition had very negative attitudes toward the tasks.
Participants who were paid the equivalent of $180 also had very negative attitudes
—that is, the money justified the action, so there was no cognitive dissonance. But
Woo00037_Ch06.indd 213
05-01-2023 18:57:54
214
FIGURE 6-14
Research shows
that attitudes can
predict behavior,
and behavior can
predict attitudes.
© rvlsoft/Shutterstock
.com
Woo00037_Ch06.indd 214
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
FPO
for participants who were only paid $1, they changed their attitude to be more consistent with their behavior and came to believe that the tasks were, in fact, fun and
interesting. In the $1 condition, participants resolved the cognitive dissonance that
resulted from the inconsistency in their attitude (the study was dull and boring)
and their behavior (I told someone it was fun and interesting) by changing their
attitude to be more in line with their behavior. For a short video clip of footage from
the original study, search on YouTube using the keywords “Festinger Dissonance”
and click on the old black-and-white clips showing participants in the study (and
hear from Festinger himself).
This basic finding from the classic cognitive dissonance peg-board experiment
can seem counterintuitive. To further illustrate how individuals change their attitudes to align with their behavior, let’s consider a second classic social psychological study about cognitive dissonance. In this study by Aronson and Mills (1959),
participants were university students who signed up for a group discussion about
the psychology of sex. But before they were allowed to participate in the discussion,
they needed to demonstrate that they were prepared for such an experiment. For
participants assigned to the “severe initiation” experimental condition, they were
asked to read aloud a list of obscene words, as well as two explicit and lurid passages
from a romance novel about a sexual encounter. In the “mild initiation” condition,
participants were asked to read aloud words that were related to sex but weren’t
embarrassing, such as virgin or prostitute. In the control condition, no initiation
was needed, and they were allowed to go straight to the group discussion.
The “group discussion” turned out to be a recorded conversation by three female students about the anatomy and biology of sexual activity in animals, and it
was designed to be uninformative and uninteresting. After listening to the group
discussion, the participants were asked to rate the conversation on nine aspects,
05-01-2023 18:57:55
CHAPTER 6: Attitudes and Behavior
215
1.6
Ratings of enjoyment of task
1.4
1.2
1
0.8
0.6
0.4
0.2
FIGURE 6-15
0
–0.2
–0.4
–0.6
Control
$1
$20
Experimental condition
Cognitive dissonance: Results
from the “pegboard” study.
Source: Festinger &
Carlsmith (1959)
including interest in the discussion, intelligence of the speakers, and level of engagement, on a scale from 0 to 15. Not surprisingly, the ratings were generally low.
But what about the ratings for participants in the mild or severe initiation? From a
cognitive dissonance perspective, who do you think would report more favorable
attitudes after listening to the discussion? The results showed that the mild initiation group was not statistically different from the control. In contrast, the severe
initiation group reported significantly more favorable attitudes (average ratings of
nearly 11 on the 0-15 scale). In essence, participants in the severe initiation condition needed to justify their embarrassing experience by creating a more favorable
attitude toward the resulting task.
3. Justify the behavior and retain the attitude. Up to this point, we’ve seen research
examples showing that individuals who find themselves in a dissonance situation
can resolve the dissonance by changing their attitude or by changing their behaviors. But dissonance can also be resolved by justifying the actions. Have you ever
found yourself in a group where someone made an offensive statement, and you
felt like you should confront the person, but the moment passed and you failed to
act? This was the context used in a series of studies conducted by Rasinski et al.
(2013). In these studies, female students participated in a study that involved working
with a partner to solve the “Deserted Island Task.” Each of the female participants
worked with a male confederate (a confederate is a person who is part of the research team but pretends to be another participant).
In the Deserted Island Task, a team of two must work together to select from a
list of potential individuals those they believe would be most helpful for surviving
on a deserted island. There is a pool of 30 possible individuals to select from, each
with different occupational information and photos. This was the task assigned to
participants during the study, and while the female participants were discussing
with their “partner” to determine the best selections, the male confederate made
a sexist comment—for example, “She’s pretty hot. I think we need more women
on the island to keep the men satisfied.” In the study, the female participants were
randomly assigned to either have an opportunity to confront the sexist confederate
Woo00037_Ch06.indd 215
05-01-2023 18:57:55
216
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
or were not given such an opportunity. In the no-opportunity condition, a buzzer
sounded immediately after the confederate’s statement, and the researcher announced that the study was complete and gave instructions to move on to the next
part. In the opportunity condition, there was a short break (10 seconds), which
allowed the participant a chance to confront their sexist partner.
Prior to subjects’ participation in the study, the researcher measured each
person’s attitude toward confrontation. In the first experiment, 22 of the 72 participants who had an opportunity to confront the sexist confederate did so. Interestingly, there was no relationship between the person’s initial attitude toward
confronting a sexist remark and actually doing it. Following the experiment, each
female participant then rated her impression of the confederate on several valenced
items: intelligent, likable, kind, smart, polite, offensive, biased, annoying. Each item
was rated from 1 = not at all to 7 = very much. The negative items were reversed, and
the scores were averaged to produce an overall favorability measure. The results
showed that for participants who had an initially strong attitude toward confronting a sexist remark but failed to do so, having an opportunity to do so resulted in
more positive evaluations of their sexist partner in comparison to not having an
opportunity. These participants found themselves in a situation where they should
have confronted the sexist remark but didn’t. That is, they had a favorable attitude
toward confronting, but their behavior was inconsistent, resulting in cognitive dissonance. To resolve the dissonance, they changed their attitude toward the sexist
partner to be more favorable (note that this is another demonstration of the classic
dissonance effect discussed earlier with the peg-board study).
But in a second study, the researchers showed that there is another pathway to
resolve dissonance in which we can justify our actions and retain the initial attitude. Following a similar procedure as the first study, the participants were offered
an opportunity to “trivialize” their lack of action—for example, “It wasn’t that important for me to tell him he was being a jerk.” The results showed that participants
who trivialized the lack of confrontation did not change their attitude toward the
sexist confederate. In contrast, those participants who did not trivialize reported
less favorable views of the sexist confederate. This shows that it is possible to retain
the original attitude by adding new cognitions that justify the inconsistency between attitudes and action.
SELF-PERCEPTION THEORY
In the previous section, we saw several examples where individuals changed their
attitudes to be consistent with their behavior. These findings were interpreted as evidence for cognitive dissonance theory, and indeed, there is considerable research
to support the dissonance explanation. But there is an alternative, and simpler, explanation for the attitudinal differences found in dissonance studies like the classic
peg-board study. Recall our discussion of self-perception theory in Chapter 4 – The
Self and Social Identity. Self-perception theory proposes that individuals come
to know their attitudes by observing and interpreting their own behavior. From
the self-perception perspective, experiencing the aversive state of dissonance is not
Woo00037_Ch06.indd 216
05-01-2023 18:57:55
CHAPTER 6: Attitudes and Behavior
217
necessary to explain these effects. Instead, individuals simply monitor their own
actions and the context in which these actions take place, then create attitudes that
are in line with these observations (Bem, 1972).
As initial evidence to support the self-perception explanation of attitude formation, Bem (1967) randomly assigned participants to read detailed descriptions of
one of the three experimental conditions in the peg-board study. After the description, participants then listened to a tape of a man (“Bob”) describing the boring
peg-turning tasks as fun and interesting, and they were told that he was either paid
$1 or $20. Participants then rated the extent to which the man found the peg-board
tasks fun and interesting. The pattern of results from the simulated experiment was
similar to that from the original, and participants rated the man in the $1 condition
as having more favorable attitudes than the man in the $20 condition. But importantly, there was no dissonance involved. The participants were merely observing
the person’s actions and drawing conclusions about his behavior.
Other studies also provide support for the self-perception theory. One line of
work has focused on emotion and the ways in which individuals come to have
a specific emotional experience. According to self-perception theory, emotion is
merely a cognitive label that an individual derives by observing their behaviors
and context. To illustrate, studies have shown that changing a facial expression can
cause changes in emotion and changes in attitudes. Laird (1974; Laird & Bresler,
1992) asked participants to contract or relax various facial muscles in a manner
that resulted in their faces forming the shape of a smile or a frown. But this was
done in such a way that the participants were unaware that they were “smiling”
or “frowning.” Yet when asked to report their mood state, participants who were
induced to put their faces in the shape of a smile reported being happier than participants who were induced to put their faces in the shape of a frown.
Finally, let’s extend this work on self-perception theory to examine how it can
explain real-world attitudes. In a series of studies by Ito et al. (2006), participants
were induced to put their faces in the shape of a smile by having them hold a pencil
in their mouths without letting it touch their lips, just their teeth. Try it. As you see,
holding the pencil causes your mouth to form the shape of a smile. The researcher
then had these participants view photographs of faces of unfamiliar people, all of
which were either Black or White individuals. This served to pair smiling with either Black or White individuals. Following the images, participants then completed
an IAT to measure implicit racial attitudes toward Blacks and Whites. The results
showed that participants who were “smiling” while viewing the images of Black
individuals showed more favorable attitudes toward Black individuals than did participants who were “smiling” while viewing images of White individuals. The basic
interpretation from a self-perception perspective is that observing ourselves smiling while viewing images of Black individuals resulted in more favorable attitudes
—“I’m smiling, so I must like individuals from this group.”
The facial-feedback hypothesis just summarized has become a controversial
topic in social psychology. On the one hand, some studies have reported strong
results, and the findings make intuitive sense. But on the other hand, a number of
researchers and labs have reported failures to replicate the finding (see Chapter 2
about the importance of replication in social psychology). In an attempt to replicate
Woo00037_Ch06.indd 217
05-01-2023 18:57:55
218
SOCIAL PSYCHOLOGY: Linking Theory, Research and Action
the basic effect, Wagenmakers et al. (2016) implemented a similar research protocol across 17 independent laboratories. In each of these labs, the researchers used
identical materials and methods, and the results were pooled. Across these 17 attempts, none of the labs produced statistically significant results. Yet despite these
discouraging results, other research has come out in support of the facial-feedback
hypothesis. Coles et al. (2019) conducted a meta-analysis of 138 studies that experimentally manipulated facial feedback and examined the impact on self-reported
emotions. Based on their review of the existing evidence, the authors conclude that
facial feedback does in fact influence emotional experience, although the effects
tend to be small and variable.
As we’ve seen, self-perception theory provides an alternative explanation for
research findings regarding cognitive dissonance. So, which is right? Although to
date, there is no definitive test showing the validity of one theory over the other,
there have been attempts to integrate the two perspectives (Fazio et al., 1977). One
possibility is that cognitive dissonance theory works best in situations where the
be...
Purchase answer to see full
attachment
