In a classic article published in the mid-1930s, Gordon Allport (1935) contended that the attitude concept was ‘the most distinctive and indispensable concept in contemporary social psychology’. While this confident assertion may perhaps be more debatable now, the study of attitudes continues to occupy the attention of many researchers.
Attitudes are predominantly a matter of affective evaluation. They represent the evaluations (positive or negative) that we associate with diverse entities, for example, individuals, groups, objects, actions and institutions. Attitudes are typically assessed through a direct inquiry procedure in which respondents are essentially asked to indicate their evaluative reaction (like-dislike, and so on) to something or someone. A number of indirect (disguised) measurement procedures have also been developed (Kidder and Campbell 1970), but these are sometimes difficult to apply and have not been widely utilized.
Some theorists contend that attitudes should not be defined solely in affective (or evaluative) terms, suggesting instead that attitudes are normally found in combination with ‘related’ cognitive and behavioural components. Thus, people who like unions will usually hold characteristic beliefs; they may believe, for example, that union activities have often been treated unfairly in the press. In addition, people with pro-union attitudes will often act accordingly, by joining a union, or by purchasing union goods in preference to those produced by non-unionized labour. Despite the plausibility of these assertions, however, they have not gone unchallenged; in particular, the relationship between attitudes and behaviour has often proven to be weak or nonexistent.
Rather than defining attitudes such that associated beliefs and behaviours are included as essential components (by definition), contemporary researchers have preferred to focus on the evaluative aspect of attitudes, to judge from the assessment procedures they have developed, and have gone on to study empirically the relationship between attitudes and beliefs and the relationship between attitudes and behaviour.
Attitudes and beliefs
A commonsensical approach suggests that our attitudes, pro or con, derive from our beliefs. For example, if we learn that a newly opened store offers excellent service, superior goods and low prices, we are likely to evaluate it positively. Advertising campaigns are often based on an implicit model of this type; they may attempt to change our beliefs about a product or institution by telling us of the good qualities it possesses, in the hope that this will ultimately influence our attitudes and buying behaviour.
While it is clear that attitudes can be influenced by changes in belief (as outlined above), there is also evidence for the reverse proposition. That is, attitudes may not only be influenced by beliefs, but they may also contribute to the things that we believe (Rosenberg et al. 1960). In one study, for example, respondents were led (through direct post-hypnotic suggestion) to accept a new position with respect to foreign aid. Subsequent inquiry indicated that these hypnotically induced attitudes were accompanied by a spontaneous acceptance of new beliefs that had not been mentioned during the induction procedure, beliefs that were supportive of the respondents’ new views. Other studies suggest that attitudes may also play a type of filtering role, influencing the extent to which we accept new information that bears on the validity of our attitudes (Lord et al. 1979).
Attitudes and behaviour
Attitudes are generally thought to influence behaviour. People who favour a given candidate or political position are expected to vote for that person, or to provide other concrete support (for example, in the form of donations), in contrast to those who hold relatively negative views. Despite the seeming obviousness of this proposition, however, many studies have found only weak, unreliable relations between attitudes and everyday behaviour. Part of the difficulty here derives from the fact that behaviour is often dependent on situational factors that may override the influence of the individual’s preferences. Despite the fact that someone holds extremely positive views towards organized religion, she may none the less be unresponsive to requests for financial donations to her church if she has recently lost her job. Similarly, a hotel clerk may override his personal prejudices and politely serve patrons of diverse ethnic origins, if this is what his job requires. On the other hand, there is now persuasive evidence that attitudes may be more substantially associated with everyday actions if we take a broader view of behaviour, tracking the individual’s reactions in a wide range of settings rather than just one. For example, although religious attitudes (positive-negative) may be weakly associated with financial contributions to the church, a more clearcut linkage between religious attitudes and religious behaviour may be observed if a composite behavioural index is employed, one that takes account of such matters as weekly religious observance, observance during holiday celebrations, saying Grace before meals, and so on (Fishbein and Ajzen 1974). Attitudes may also be effectively related to overt actions if they are action-oriented and are measured with appropriate specificity. Thus, church donations may be related to people’s attitudes toward the concrete act of ‘donating to the church’, as contrasted with their general attitude towards ‘organized religion’.
One of the most firmly established phenomena in contemporary attitude research is the fact that behaviours may have a causal impact on attitudes, rather than simply reflecting the actor’s previously held views. This proposition has been supported in a wide range of experiments. In a classic study by Festinger and Carlsmith (1959), some respondents were led to describe a certain laboratory activity as ‘interesting’, despite the fact that they actually regarded it as rather dull. People who had enacted this form of counter-attitudinal behaviour for a modest (one-dollar) incentive subsequently rated the dull laboratory task in relatively favourable terms, compared to those who had not been required to produce counter-attitudinal statements. Other researchers have employed a procedure in which a person who was supposed to be ‘teaching’ something to another seemingly punished the learners with electric shocks whenever they made an error. Subsequent enquiry revealed that people who had served as ‘teachers’ in this type of situation became increasingly negative to their ‘pupils’ as a consequence.
The continuing vitality of the attitude construct may derive, in part, from the seemingly universal importance of evaluation (Osgood 1964). We are apparently disposed to respond evaluatively to the people, objects, events and institutions that we encounter. These evaluative (attitudinal) reactions, their origins, correlates and consequences, continue to constitute a fertile domain for academic and applied research.
Melvin Manis
University of Michigan
References
Allport, G.W. (1935) ‘Attitudes’, in C.Murchison (ed.) A Handbook of Social Psychology, Worcester, MA.
Festinger, L. and Carlsmith, J.M. (1959) ‘Cognitive consequences of forced compliance’, Journal of Abnormal and Social Psychology 58.
Fishbein, M. and Ajzen, I. (1974) ‘Attitudes toward objects as predictive of single and multiple behavioral criteria’, Psychological Review 81.
Kidder, L.H. and Campbell, D.T (1970) ‘The indirect testing of social attitude’, in G.I.Summers (ed.) Attitude Measurement, Chicago.
Lord, C.G., Ross, L. and Lepper, M.R. (1979) ‘Biased assimilation and attitude polarization: the effects of prior theories in subsequently considered evidence’, Journal of Personality and Social Psychology 37.
Osgood, C.E. (1964) ‘Semantic differential technique in the comparative study of cultures’, American Psychologist 66.
Rosenberg, M.J., Hovland, C.I., McGuire, W.J., Abelson, R.P. and Brehm, J.W. (1960) Attitude Organization and Change, New Haven, CT.
Further reading
McGuire, W.J. (1969), ‘The nature of attitudes and attitude change’, in The Handbook of Social Psychology, 2nd edn, vol. 3, Reading, MA.
Petty, R.E. and Cacioppo, J.T. (1981) Attitudes and Persuasion: Classical and Contemporary Approaches, Dubuque, IA.
