Social Psychology

by Paul Seager

  It is generally agreed that attitude content can be split into three components – the ABC of attitudes:

  1 Affective

  2 Behavioural

  3 Cognitive

  Key idea: Attitude content

  This refers to the affective, behavioural and cognitive components of an attitude.

  The affective component refers to how we feel about something, and the emotions that it evokes. The behavioural component refers to how we act (or have acted) with regards to the object, and our experiences with it. The cognitive component refers to how we think about an object, the beliefs that we have with regard to it, and also the attributes we link with an object. For example, my attitude towards cricket is that I enjoy watching it as I find it quite soothing (affective element), I have played it in the past quite successfully (behavioural element) and believe it to be an intriguing game which not only requires physical skills but psychological ones too (cognitive element). This is an example of where all of the components of an attitude are in one direction, e.g. positive: this is referred to as the valence of an attitude.

  However, some attitudes can be mixed. To take a simple example, I might think that having some knowledge of classical literature would make me a more rounded, better educated person (positive cognitive) but feel that life is too short to spend so much of my time on such a pursuit (negative affective) especially as when I have tried in the past to read such books they just send me to sleep (negative behaviour). If I went on to list many more positive aspects along with many more negative aspects of reading classical literature, this would represent an ambivalent attitude. This then leads us to consider the question of the attitude structure.

  There seem to be two different views on the structure of attitudes:

  1 One-dimensional view: this suggests that attitudes are measured on a continuum, one end of which denotes positivity and the other end denotes negativity. Completely positive or negative attitudes would seem to be accounted for by this view, as would situations where an attitude comprises many positive but one or two negative views (or vice versa).

  2 Two-dimensional view: this suggest that attitudes should be plotted on a two-axis graph, where one dimension represents the amount of positive views with regard to the attitudinal object, and the other dimension represents the amount of negative views. Ambivalent attitudes seem better represented by this view.

  Key idea: Attitude structure

  This refers to the positive and negative dimensions of an attitude, and how they are distributed across the affective, behavioural and cognitive components.

  The two-dimensional approach would seem to give a richer, more complete view of attitudes: for example, imagine an individual was asked to rate their attitude towards cricket on a nine-point scale (see attitude measurement below) where 1 indicated a negative view, 9 indicated a positive view and 5 was the mid-point: this would represent a one-dimensional view. If the individual circled the ‘5’, what would we be able to infer about their view of cricket? It could indicate that they have never played or watched it, therefore they are using ‘5’ to indicate that they have no positive or negative feeling; alternately, they might have lots of positive and negative views towards the game, and to represent this balance, they might choose ‘5’ as a compromise position. A two-dimensional representation would give us a clearer view of which of the two positions was represented by the individual’s attitude.

  Whilst everyone has many attitudes, some are certainly held more strongly than others. Attitude strength can signal key information, such as how important an attitude is to an individual and the degree of certainty with which they hold a particular attitude. Certainly my attitude towards cricket is a lot stronger than my attitude towards classical literature. In general, it has been found that strong attitudes differ from weak attitudes on a number of dimensions:

  • Durability: strong attitudes tend to be more temporally stable: they last longer (I’ve liked cricket since about the age of 15).

  • Resistant to change: strong attitudes are less likely to crumble in the face of persuasive messages (no one has been able to persuade me that test match cricket is boring).

  • Influence: individuals are likely to pay greater attention to information which is related to strong attitudes than they are to information related to weak attitudes (I’m more likely to read the cricket section of the sports pages than I am to read the horse racing section).

  • Behaviour: strong attitudes tend to influence behaviour far more than weak attitudes (if I have the choice between watching cricket on television or reading a work of classic literature … well, I think you know the answer to that one!).

  In order to understand why people hold different attitudes, it is necessary to understand the attitude function, and for most people they serve multiple purposes. According to Smith et al. (1956), attitudes have three primary functions:

  1 Object appraisal: they help us to weigh up the positive and negative features of objects. This will help us to avoid harmful things and embrace the positive ones. It is generally considered to be the most important function of an attitude.

  2 Social adjustment: they help us to identify others who we might like, and help us to avoid those we dislike.

  3 Externalization: they protect us against things that might cause us some degree of internal conflict.

  Others have proposed slightly different functions of attitudes, some of which bear a similarity to the three listed above, and some which add to them. For example, Katz (1960) argued that attitudes help us to express who we are and help to define us in terms of our self-concept; through the attitudes we hold, and share externally, others are able to learn about us. Some attempts have been made to measure the functions of attitudes more explicitly, such as the ‘attitudes functions inventory’ (Herek, 1987). However, some critics argue that such measurement tools are not accurate as individuals may not know themselves what functions their attitudes serve.

  Key idea: Attitude function

  This refers to the different purposes that an individual’s attitudes serve.

  Having looked at what an attitude is, and at the different aspects of attitudes, the next question to consider is how we might measure them.

  Measuring attitudes

  Attitudes are not tangible objects, instead they are buried somewhere inside an individual’s head. This makes measuring them quite difficult (but that has never stopped social psychologists before). Generally speaking, there are two main ways of measuring attitudes:

  1 Explicit methods: these ask direct questions about an attitude; an individual will be aware that this is happening.

  2 Implicit methods: these tap into attitudes indirectly; an individual is usually unaware that their attitudes are being assessed.

  Explicit methods typically use self-report questionnaires which ask an individual to report on their opinions about an attitudinal object. A typical questionnaire of this nature will use a ‘Likert’ scale, with some items being reverse scored, in order to assess an individual’s attitude towards a particular attitudinal object, such as a political party or a brand of bread (see Spotlight below). This method is quite common, though limited in that the construction of questions ties a questionnaire to a specific object.

  Other direct methods of measuring attitudes include the ‘semantic differential approach’, which has the advantage of being able to compare attitudes across topics. It simply uses a number of adjectives (such as good/bad; positive/negative) as anchor points on a scale. Individuals then mark a point on the scale to reflect their views of the attitudinal object. Using this method, it would be possible to compare an individual’s attitude towards both cricket and classical literature more easily than using a Likert scale, which would require two specially constructed scales.

  However, there are a number of problems with using direct methods of attitudinal measurement. For example, an individual might not actually be aware of what their attitude towards the object actually is. Add
itionally, great care must be taken with question development as slight differences in the way a question is worded can lead to different responses. However, the biggest issue is probably due to social desirability, which means that instead of giving responses that reflect their real opinions, an individual will more likely give a response that presents them in the best possible light (whilst this wouldn’t necessarily be a problem with our cricket survey, it could well be a problem with more contentious issues such as immigration).

  In an attempt to avoid some of the problems encountered with direct measures, a number of implicit (indirect) methods have been formulated. One such method is evaluative priming, and is based on the assumption that attitudes are associations between objects and evaluations, and the stronger the association, the more quickly it can be retrieved from memory. It also assumes that if an individual encounters the object, the evaluation will be triggered. To take such a test, an individual would be seated in front of a computer, and shown the name of the attitudinal object (referred to as the ‘stimulus prime’). This is quickly replaced by an evaluative adjective (such as ‘disappointing’ or ‘exciting’). The individual’s task is to classify the adjective as quickly as possible as representing either a good or bad thing.

  The speed with which the judgement is made is the key. In general, if we are trying to make sense of something, we find it easier to do so if we have recently seen something that is similar. Thus, if the individual is exposed to a stimulus prime that they like (e.g. cricket), they will be quicker at classifying a positive adjective (e.g. ‘exciting’) as good, but slower at classifying a negative adjective (e.g. ‘disappointing’) as bad. Similarly, if the stimulus prime was something they disliked (e.g. classical literature), they would be slower to classify a positive adjective, but quicker to classify a negative adjective.

  Spotlight: Measuring attitudes directly

  Questionnaires typically use a Likert scale to assess attitudes towards an object. For example, to measure attitudes towards cricket, researchers might construct a scale as follows:

  The following questions are statements about the game of cricket. Please use a number from the following scale to reflect your personal opinion for each statement.

  1 = Strongly disagree

  2 = Disagree

  3 = Uncertain

  4 = Agree

  5 = Strongly agree

  1 Cricket is a game of skill.

  2 Watching cricket helps me to relax.

  3 Cricket holds no interest for me.

  4 Cricket is a game of brute force and ignorance.

  Each of the four questions would be answered using the 1–5 scale. Questions 1 and 2 reflect positive views towards cricket, and thus individuals giving low scores are reflecting a negative attitude; however, as questions 3 and 4 are reflecting negative attitudes, they would need to be reverse scored in order to provide a consistent view, thus anyone giving a ‘5’ to question 3 or 4 (a negative attitude) would actually have their score changed to 1 (a score of 4 would be changed to 2). The scores from the four questions would then be added up to give an attitude score between 4 and 20; in this case, higher scores would reflect a more positive attitude towards cricket, and lower scores a more negative view.

  However, using the semantic differential approach, attitudes towards cricket could be assessed by using a number of paired adjectives as follows:

  Cricket

  Bad :___:___:___:___:___:___:___:___:___: Good

  Negative :___:___:___:___:___:___:___:___:___: Positive

  Dull :___:___:___:___:___:___:___:___:___: Interesting

  Using this method, attitudes towards other attitudinal objects (such as classical literature) could easily be measured and compared with one another.

  Tests based on evaluative priming have been modified to test less desirable attitudes such as racism (imagine showing a picture of a white or black person as the stimulus prime, and then showing the word ‘intelligent’). The ‘implicit association test’ (IAT) is another such indirect measure of attitudes, and has been used to measure prejudicial attitudes (see Chapter 13).

  As well as direct and indirect methods, there has been some attempt to measure attitudes through an individual’s physiological responses. For example, pupil dilation and galvanic skin response (essentially a skin sweat response) have both been linked with arousal, and the idea is that a strong arousal response would be a good indicator of the strength of an attitude. However, whilst there is some merit in this view, it is not always possible to determine the valence (positive or negative) of the attitude through such measures.

  More recent research has used different types of brain scans (e.g. functional magnetic resonance imaging) to attempt to measure attitudes. However, these findings have not always produced good results; for example, voters were shown pictures of American presidential candidates, and according to the results of brain scans, Mitt Romney showed potential as a candidate, but Barack Obama showed neither positive nor negative signs. Add to this lack of predictive validity, the cost of individual brain scans (probably in the region of several hundred pounds a time), and it is easy to see why this method could be a non-starter.

  Do attitudes predict behaviour?

  The question of whether attitudes can predict behaviour is an interesting one. An early study by LaPierre (1934) served to illustrate this point. He was travelling across America with a young Chinese couple at a time when there was a strong anti-Asian prejudice, and was concerned about whether or not they would be able to find places to stay and eat. However, only once in over 250 establishments were they refused service which he found quite surprising. Upon the conclusion of his travels, LaPierre wrote to each of the establishments to ask whether or not they would be prepared to serve Chinese visitors; he found that only one establishment said that they would. Given his experiences on his travels, this was presented as evidence that individual attitudes were not good predictors of behaviour.

  However, subsequent research, which has been conducted a little more rigorously than the study by LaPierre, suggests that the relationship between attitudes and behaviour may not be quite so simple. In fact, if the conditions are right, then attitudes can indeed predict behaviour, but this is contingent on a number of factors which include:

  • Strength of attitude: stronger attitudes are much more likely to predict behaviour than weaker ones. However, on some occasions, a given behaviour can lead to the formation of a weak attitude (suggesting that the relationship between attitudes and behaviour is not always uni-directional).

  • Function of attitude: attitudes used to express values are much more likely to predict behaviour.

  • Topic of attitude: attitudinal topic can have an influence on behaviour. For example, research suggests that political attitudes are a better predictor of behaviour than attitudes towards blood donation.

  • Individual differences: the personality of an individual has been found to affect the extent to which attitudes predict behaviour. For example, they are a better predictor for people scoring low in ‘self-monitoring’ than high scorers, but a better predictor for individuals scoring high in the trait of ‘need for cognition’ than lower scorers.

  A number of different models have been formulated to help show how attitudes can predict behaviour, and probably the most well-known of which is Fishbein and Azjen’s (1975) ‘theory of reasoned action’. This model helps to explain the role that attitudes play in deliberate behaviour, and is based on the assumption that an individual’s intention to act is the main determinant of their behaviour. Intention is influenced both by attitude (towards the behaviour) and subjective norms. An attitude in turn is affected by the anticipation that any behaviour will lead to a desired outcome, and also by the value that is attached to the desired outcome. Subjective norms refer to an individual’s beliefs about how important others will view the behaviour. These norms are affected by an individual’s beliefs about how they think others expect them to behave and how motivated the indiv
idual is to live up to these expectations.

  There was much support for the model, but it quickly became clear that a third factor was also capable of influencing both an individual’s intention to act and their behaviour – namely perceived behavioural control. This led to a revision of the theory, with the outcome being the ‘theory of planned behaviour’. Perceived behaviour control refers to whether or not an individual is actually able to carry out the behaviour, regardless of their attitude and any subjective norms. For example, an individual might feel strongly about giving their old clothes to charity (attitude), and believe that others would approve of their behaviour (subjective norm), but unless they had some means actually to donate their clothes (such as a nearby charity shop and a means of getting there), the behaviour would not occur.

  The model suggests that perceived behavioural control moderates behaviour in two ways: it impacts on an individual’s intentions by assessing their confidence in their ability to perform the actions (i.e. do they have the resources needed to perform the behaviour?), and it impacts on an individual’s behaviour by objectively assessing whether the action can realistically be performed (whilst an individual might believe they can perform an action, their perception might not be an accurate one).

  Whilst the model certainly has the support of much empirical research, it is not without limitations: for example, it is not always clear as to which intention-affecting component exerts the greatest influence (or how this might change across situations), and it doesn’t tackle the issue of how intentions get translated into action. Nevertheless, the model does give us an interesting insight into the attitude-behaviour question by suggesting both that attitudes are only one of a number of factors that might influence our behaviour, and that their effects are indirect ones.