attitudes

attitudes, attitude research Variously defined as an orientation towards a person, situation, institution, or social process, that is held to be indicative of an underlying value or belief; or among those who insist that attitudes can only be inferred from observed behaviour, as a tendency to act in a certain (more or less consistent) way towards persons and situations. Milton Rokeach's much-quoted definition (Beliefs, Attitudes and Values, 1976) embraces both tendencies and describes an attitude as ‘a relatively enduring organization of beliefs around an object or situation predisposing one to respond in some preferential manner’.

Social psychologists and sociologists have invested a great deal of effort in the measurement of attitudes, opinions, and views on society at large; on relationships and events within it; and on the identification and measurement of underlying values, which are less volatile, more deeply held ‘prejudices’. Attitudes are studied both as a substitute for measuring behaviour directly and because they are (sometimes) assumed to predict behaviour. Some social scientists treat them as important variables in their own right, key features of the individual, as reflected for example in the so-called authoritarian personality.

The sheer volume of attitudinal research is not difficult to explain. Consider, for example, the phenomenon of race discrimination. It is not easy to observe instances of discrimination, and isolated incidents, while illustrative, may not be representative. The alternative in surveys is to ask people to report their behaviour, but this runs into difficulties with situations that have never arisen, or are purely hypothetical. The other approach is then to collect attitudinal data on people's predispositions and stated values, the advantage being that the questions seem to be appropriate for everyone.

In reality, however, many people do not have well-developed or even superficial opinions on topics that may interest the sociologist. Some would argue that the idea of attitudes is closely tied to the culture of Western industrial society, in which citizens are regularly invited to express their views on public issues, both directly and through the ballot box. What is certain is that attitude scales developed in Western societies do not function in the same way in other cultures. Even the standard simple job satisfaction question attracts a different pattern of response as soon as it is used beyond the confines of Western industrial societies—in Japan for example. There is some debate about the ethnocentricism and broader cross-cultural validity of many attitude scales that have been developed over the past eight decades.

At the simplest level, attitude questions invite people to agree or disagree, approve or disapprove, say Yes or No to something. More sophisticated techniques for measuring attitudes (treated under separate headings in this dictionary) include the well-established and easy to use Likert scale, the Thurstone scale, Osgood's semantic differential scale, the Bogardus social distance scale (in which attitudes are equated with hypothetical behaviour), and Guttman scales. A huge variety of personality tests, attitude and aptitude scales have been developed in the United States and Europe for commercial use by employers and recruitment agencies, as part of the staff selection process. Attitude scales of various types are sometimes used in opinion polls, occasionally in simplified form. Attitude research shades into studies of reported behaviour, sociometric scales, the sociology of knowledge, research on motivations, preferences, aims and objectives, which are also causally linked to behaviour, and the whole range of social psychological research.

One of the longest-running disputes in social-scientific research concerns the relationship between attitudes and action. What are the behavioural implications of holding particular attitudes? This debate, which has taken place mostly among social psychologists, has culminated (after some six decades of research) in ‘general agreement that attitude, no matter how assessed, is only one of the factors that influence behaviour’ (see Icek Ajzen and and Martin Fishbein , Understanding Attitudes and Predicting Social Behavior, 1980
). However, this is as far as the consensus goes, since there is then widespread disagreement, both about where attitudes stand in this list of factors, and how they relate to the many other variables in the equation. After exhaustively reviewing the literature, Richard Eiser (Social Psychology: Attitudes, Cognition and Social Behaviour, 1986) was able to conclude only that ‘attitudes, in short, have behavioural implications’, and that ‘the question of which specific behaviours are implied by a particular attitude, however, will depend on circumstances, and is therefore an empirical one’.

For a useful summary of the major contributions to this debate, together with a discussion of other salient issues in attitude research, see Richard Eiser and and J. van der Pligt , Attitudes and Decisions (1988)
. See also EQUAL APPEARING INTERVALS; PROTESTANT ETHIC THESIS.