Chapter 6: Attitudes and Social Behaviour
What Are Attitudes?
Geoff Keller‟s (president of „Attitude Is Everything, Inc.‟) definition of an attitude: a general perspective
on life, an outlook that can be positive (leading to success) or negative (leading to failure); corresponds
closely to the dimension of dispositional optimism (discussed in Chapter 5)
Social psychologists have a much more specific meaning for the term attitude.
Attitudes: Evaluations of Targets
Attitude: an individual‟s evaluation of a target (can be an object, an issue, a person, a group, a behaviour,
or any other identifiable aspect of the environment (e.g. a colour, an emotion))
By calling it an evaluation, theorists mean that an attitude is a good-bad judgement: it represents the
individual‟s overall assessment of whether a particular target is positive or negative.
Researchers must infer attitudes from individuals‟ observable responses as attitudes cannot be seen
Although it is true that people can have an attitude toward almost anything, social psychologists have
been interested mainly in attitudes that are directed at important targets, such as controversial issues,
ethnic groups, and consequential behaviours.
Three Parts of Attitudes
Attitudes can come from sources: emotional reactions, cognitive information, and past behaviour.
Whether an individual evaluates a target positively or negatively will depend on three things:
1) How the object makes them feel
a. Targets that arouse negative feelings and emotions (e.g. snakes, hypodermic needles) are
more likely to generate unfavourable attitudes than are targets that arouse positive
feelings and emotions (e.g. puppies, chocolate).
2) The person‟s beliefs about the object
a. Targets that are known or believed to possess negative characteristics (e.g. criminals,
cholesterol) are more likely to generate unfavourable attitudes than are targets that are
known or believed to possess positive characteristics (e.g. medical doctors, healthy
3) The person‟s previous actions toward the object
a. Targets toward which someone has behaved negatively in the past (e.g. enemies, weeds)
are more likely to be seen as disliked than are targets toward which someone has behaved
positively in the past (e.g. friends, kittens).
Two-way relation between attitudes and behaviour: past behaviours influence current attitudes, and
current attitudes influence future behaviour. Attitudes toward certain targets depend mostly on people‟s feelings toward those targets (e.g. attitude
towards blood donation depends on one‟s fear of donating blood), whereas attitudes toward other targets
depend mostly on people‟s knowledge and beliefs (e.g. attitude toward controversial social issues depend
on one‟s agreement with arguments supporting each side of the issue).
Ambivalent attitudes: evaluations of targets that include both positive and negative elements; this conflict
among the attitudinal elements is experienced as unpleasant
A key implication of ambivalence is for the consistency of behaviour – what kind of behaviour does an
ambivalent attitude produce? Is it constant or variable?
Ambivalent attitudes can lead to different behaviour over time because either the positive or the negative
elements about the target may come to mind at a particular point, and whichever type of element is
dominant will drive behaviour (accessibility).
Attitudes that are low in ambivalence (all positive or all negative) will not produce such variable
Explicit Versus Implicit Attitudes
Explicit attitudes: those that people can report consciously
Implicit attitudes: an individual‟s automatic evaluative response to a target, which can occur without
awareness; spontaneous good-bad response that cannot be consciously controlled
Typically, implicit attitudes conform to explicit attitudes (they match); however, inconsistency between
the two does occur – people may not always realize that their implicit and explicit attitudes toward a
Perceptions of Others’ Attitudes
There is a common structure to people‟s perceptions of others‟ attitudes. Two dimensions are most
1) Liberal vs. conservative
2) Traditional vs. novel (or radical)
Thus, perceptions of other people‟s attitudes tend to be guided by consideration of the extent to which the
others are liberal or conservative and traditional or innovative.
Why Do We Evaluate?
Recognition, per se, is not very informative unless the memory system also triggers some sort of
evaluation of the object.
Object-appraisal function: attitudes give the individual a quick assessment (appraisal) of whether targets
are likely to be helpful or hurtful; most basic function of/principal reason for attitudes The fundamental goal of object appraisal can be served by simple, affective responses that do not rely on
complex cognition, causing attitude-development to generalize beyond humans.
Values: broad, abstract standards or goals that people consider to be important guiding principles in their
People‟s values are related to their attitudes toward specific issues (e.g. religious individuals might adopt
specific positions on issues such as birth control or gay marriage to display their support for their faith);
the attitudes serve, in part, symbolic functions for the holders
Value-expressive function: attitudes allow people to convey an identity that connects them to some
groups and makes them distinct from other groups
Testing the Functions of Attitudes
How can an investigator determine whether a particular attitude fulfills an object-appraisal or a value-
The motivations underlying object-appraisal attitudes differ from those underlying value-expressive
Sharon Shavitt‟s experiment:
Idea: attitudes toward a particular object may fulfill the same function for almost everyone
Two examples of attitudes she studied: attitudes toward coffee and attitudes toward perfume
Shavitt proposed that one of these attitudes typically fulfills an object-appraisal function, and the other
typically fulfills a value-expressive function.
Object-appraisal attitudes give the individual a quick evaluation of the target, whereas value-expressive
attitudes tell other people about the individual‟s identity or values.
Shavitt hypothesized that attitudes toward coffee typically fulfill an object-appraisal function: people
either like or dislike the taste of coffee (as well as its dose of caffeine) and that attitudes toward perfume
often fulfill a value-expressive function: many people purchase a particular brand of perfume because it
projects a desired image or because it is promoted by a beautiful model or movie star with whom they
Shavitt asked participants to write down thoughts about their attitudes toward a particular target (e.g.
coffee, perfume) and to explain why they felt that way. These thoughts were later examined by judges
who recorded how often the participants mentioned specific themes.
When participants described an object-appraisal attitude (e.g. toward coffee), they were likely to mention
positive or negative features of the object. When participants described a value-expressive attitude (toward perfume), they were likely to mention
their values, their identity, and what the object communicated to others.
Shavitt tested the implications of attitude functions for the effectiveness of persuasive messages – ads.
If attitudes toward coffee give people a quick evaluation of this target, how should an ad be constructed to
promote a new brand of coffee effectively?
Shavitt proposed that the most effective strategy would be to focus on the positive features of the coffee
and the rewards it will bring. But if attitudes toward perfume reflect individuals‟ identities and desired
images, how should an ad for a new brand of perfume be constructed? Shavitt proposed that the most
effective strategy would be to focus on the desirable impression the perfume will make on others.
Shavitt created two different versions of ads for fictitious brands of coffee and perfume.
One version stressed the rewards provided by the product (“The delicious, hearty flavour and aroma of
Sterling Blend coffee come from a blend of the freshest coffee beans” or “The fresh, floral scent of
Cadeau perfume comes from a balanced blend of oils and essences”).
The second version emphasized how the product created a particular image in others‟ minds (“The coffee
you drink can reveal your rare, discriminating taste” or “Cadeau perfume is the sophisticated scent that
tells people that you are not one of the crowd”).
The coffee ad focusing on rewards generated more interest than did the coffee ad focusing on image. The
perfume ad focusing on image generated more interest than did the perfume ad focusing on rewards.
Thus, the ads were more effective when they were consistent with the function fulfilled by the attitude:
object-appraisal attitudes responded to information about rewards, whereas value-expressive attitudes
responded to information about image.
Validity refers to whether a measure actually assesses what it is supposed to assess (how valid a measure
is), and reliability refers to whether participants‟ scores on the measure are stable and free from “random”
fluctuations (how reliable the scores are).
Self-Report Measures of Attitudes
Because people are usually aware of their attitudes, it seems sensible to ask them directly to report their
evaluations. Most attitude measurement techniques are self-report in nature. Items are administered either
in a paper-and-pencil questionnaire or on a computer. Respondents indicate their attitudes by circling a
number or word on a response scale or by placing an X along a response dimension.
o Most used technique to measure attitudes over the past 70 years o Respondents read a number of statements, each of which expresses a clear position (pro
or con) on an issue, or a clear attitude (favourable or unfavourable) toward a target.
o Respondents are then asked to indicate their agreement or disagreement with each item
(on a scale from „strongly disagree‟ to „strongly agree‟ which is equivalent to 1-5 (with
some reverse scoring present)).
o Higher sums of scores reflect the same direction of attitude (e.g. higher sums reflect a
Relatively easy for researchers to construct
Clear and simple for respondents to complete
Have been shown to produce reliable scores
Semantic Differential Scales
o Respondents are asked to rate an attitude object on several evaluative dimensions (such
as good-bad, favourable-unfavourable, effective-ineffective, and fair-unfair).
o The opposing adjectives appear on either end of a 5-point response scale, and respondents
are told to put an X somewhere on the scale to indicate his/her evaluation.
o Higher sums of scores reflect the same direction of attitude (e.g. higher sums reflect a
Easy for researchers to construct
Straightforward for respondents to complete
Assesses evaluations very directly, because participants rate the attitude object on
dimensions that are explicitly evaluative (including the fundamental dimension
underlying attitudes – the good-bad dimension)
o Designed to assess public opinion about an issue, event, or group
o Sometimes, the survey researcher wants to be able to generalize the findings to a larger
population (e.g. all adults in Canada), in which case the sample must be representative of
o On other occasions, survey researchers simply want a “snippet” of public opinion and
will not present their findings as necessarily applicable to larger populations.
o Most opinion surveys contain just one or two items on a particular issue, and responses
are often limited to “yes” or “no” (perhaps also “undecided”).
Useful for gathering information about public opinion, but are rarely employed
Measurement Technique Key Features Advantages and Disadvantages
Likert-type scale Respondent rates agreement or Relatively easy to construct
disagreement with attitude Clear and simple to answer
statements Reliable scores
All statements are clearly
favourable or clearly
Researcher must identify
nonvalid items and eliminate them
Semantic differential scale Respondent rates attitude object Very simple to construct
on evaluative dimensions Clear and simple to answer
All dimensions reflect the good- Very direct measures of
bad dimension evaluations
Score is sum of respondent‟s
Opinion survey Respondent answers just one or Very simple to construct
two items on each issue Useful for gathering information
Responses are usually yes, about public opinion
undecided, or no Usually not detained enough for
Researchers sometimes obtain a use in psychological research
Problems with Self-Report Measures
o All of the described techniques rest on at least two assumptions
People know what their attitudes are
They will report those attitudes honestly
o If researchers are interested in implicit attitudes, they cannot use self-report measures.
o People reporting their attitudes honestly is questionable in some cases – often one
position on an issue is more socially desirable than other positions (topics that can be
influenced by social desirability are in the domain of ethnic attitudes, attitudes towards
illegal activities, attitudes towards harmful behaviours, and attitudes toward helpful but
costly/time-consuming actions (e.g. donating, volunteering)).
o Given that people want to appear likeable, moral, and competent, they may be tempted to
shift their answers on attitude scales in the direction of the socially desirable position.
o They typically do not yield a clear and easy way to measure the ambivalence of an
individual‟s attitude (an attitude that includes both positive and negative elements).
To measure ambivalence, some researchers have asked respondents to rate the
target on both positive and negative scales separately – for instance, participants
might rate the target on a scale from „not at all good‟ to „extremely good‟ and on
another scale from „not at all bad‟ to „extremely bad‟; in this way, participants
can indicate that they have mixed reactions to the target.
Nonverbal Measures of Attitudes
Nonverbal measures of attitudes do not rely on participants‟ ability or willingness to report their attitudes,
and they may provide better assessments of people‟s unconscious, affective responses to objects than self-
report methods (which reflect people‟s conscious, often primarily cognitive, evaluations). However, these
measures are often difficult to obtain and may not be as sensitive for assessing explicit attitudes as self-
o Some researchers have used participants‟ overt behaviour to infer their attitude toward an
object. o Behavioural measures of attitudes are usually unobtrusive measures: participants usually
do not realize that their attitudes are being assessed.
o By being unobtrusive, behavioural measures reduce problems of self-presentation and
o Unfortunately, it is often difficult to design a behavioural measure of an attitude – for
instance, how could attitudes toward nuclear power plants be inferred from behaviour?
o Also, using behaviour to measure attitudes assumes that there is a strong, inevitable link
between attitudes and actions, which is not always true.
o One category of physiological reactions that has received attention is symptoms of
arousal (such as heart rate and blood pressure).
o Some researchers have found that exposure to negative or disliked objects increases
arousal as measured by these symptoms (exposure to negative object = threat = arousal
prepares one for „fight or flight‟ response).
o It appears that although heart rate and blood pressure may sometimes indicate the
intensity of people‟s feelings about a target, arousal symptoms are poor at distinguishing
between positive and negative evaluations (very serious limitation).
o Facial electromyography (facial EMG): a procedure for measuring muscle contractions in
the face that may be sensitive to positive vs. negative responses to a stimulus (specific
patters are associated with each)
o There is some evidence that the magnitude of people‟s eye-blink response to a puff of air
directed at their eye while they are looking at a target object reveals their affective
o However, obtaining these measures is a complex and time-consuming procedure.
o Facial EMG appears to be sensitive to emotional reactions, but less sensitive to
evaluations that lack a strong affective component (e.g. attitudes that are based primarily
on cognitive beliefs).
o It is possible for people to deliberately alter or inhibit some of their facial responses.
o Since responses can occur without awareness, researchers have designed procedures that
assess the extent to which people have an automatic positive or negative response to an
o Implicit Association Test (IAT): most common reaction time procedure; requires
participants to complete two sorting tasks as quickly as possible – in one task, the target
of the attitude must be sorted into the same category as some “good” objects, and in the
other task, the target must be sorted into the same category as some “bad” objects
Basic idea: if participants complete the task in which the target is associated with
“good” things more quickly than the task in which the target is associated with
“bad” things, they are assumed to have a positive implicit attitude toward the
target and vice versa
Responses on the IAT are influenced by idiosyncratic factors, such as recently
generated thoughts by the respondents. Measurement Technique Key Features Advantages and Disadvantages
Behavioural measures Observe respondent‟s actions Unobtrusive (respondent
toward attitude object unaware)
Favourable actions (e.g. approach Not possible for all attitude
object, smile at object) are objects
assumed to reflect favourable Assumes inevitable link between
attitudes attitudes and behaviour
Physiological measures Assess respondent‟s May reflect intensity but not
physiological reactions to object direction of attitude
Examples include arousal May not be very sensitive
symptoms and facial EMG
Implicit measures Respondent‟s reaction times are Respondent cannot easily distort
used to infer automatic responses answers
Example is the Implicit Shown to predict spontaneous,
Association Test (IAT) nonverbal reactions to attitude
Implicit attitudes are assumed to object
influence the speed with which Time-consuming to obtain
the attitude object can be paired
with good or bad things
How Do Attitudes Form?
Three sources – affect, cognition, and behaviour – can each contribute to the development of an
Affective Sources of Attitudes
When an object, event, or action consistently produces positive feelings or pleasurable biological
responses, we will form a favourable attitude