In our experience, conventional views of what communication is about and how it works offer little practical guidance on how to create useful communication artifacts. This article is a critique of one of those views—the idea that communication can change a thing called our ‘attitude’—and offers an alternative approach. First published in Communication News 5 (2) September/October 1992.

Common sense suggests that you can use communication campaigns to change people’s attitudes and that these changes in attitude will lead to changes in behaviour. It sounds sensible. But you may decide, once you are familiar with the history of attitude research, that there is little place for attitude or its measurement in communication campaigns.

Origins of ‘attitude’

The term ‘attitude’ as we use it today is relatively recent. In the 19th century the word ‘attitude’ more than likely referred to somebody’s posture or bearing, as in the phrase ‘striking an attitude’. We still use the term ‘attitude’ in this way to describe the orientation of aircraft.

Psychologists in the early part of the 20th century were interested in exploring how people internalised relatively stable and enduring social values and how these values influenced their behaviour. At the same time political scientists were taking an interest in the issue of public opinion. To try and articulate these complex notions, psychologists hit on what they thought might be a productive analogy; just as a body can lean or alter its attitude, so by analogy, people could somehow mentally lean towards one point of view or another. This metaphorical use of the term ‘attitude’ quickly caught on, and by the 1930s it had become a central concept in the field of social psychology. Today, this metaphorical notion of attitude—as something inside our heads—is more commonly applied than its original sense of physical orientation or posture.

Ideas are only sometimes real

Like many ideas in the history of science, the idea of attitude came to be treated as something real and substantial before it was tested. What started out as a useful idea to advance scientific research soon became a kind of social ‘fact’. It seemed such an apt analogy that many people soon forgot that it was only an analogy and started to believe that attitudes in a psychological sense really existed.

There are many ideas in the history of science that have had a similar progression, both successful and unsuccessful. Gravity is an example of a successful idea. At first it was just a way of explaining the motion of bodies in space. Gravity itself could not be observed directly; it could only be inferred to exist from its effects. But over time the overwhelming weight of evidence and the sheer explanatory power of the idea have convinced us that this invisible unobservable force does exist.

Lavoisier decomposition air

Antoine Lavoisier’s phlogiston experiment.

There are other ideas, such as phlogiston, which have been less successful. Early 18th century chemists thought phlogiston was a substance that was ‘lost’ during burning or rusting. The evidence, however, was contradictory. Some objects got lighter, some heavier following burning. Eventually, the discovery of oxygen as the element responsible for combustion resolved the contradiction, making the idea of phlogiston irrelevant.

Science is full of similar cases where the cumulative weight of evidence or profound changes in ways of thinking make some ideas untenable or irrelevant.

The evidence against attitude

We are now at a point historically when the cumulative evidence is not running in favour of the idea of ‘attitude’. In summary, the major findings are:

  • Communication does not result in predictable changes in attitude.
  • Changes in attitude do not necessarily result in changes in behaviour.
  • Measurements of attitude do not necessarily predict behaviour.
  • There is no general agreement among researchers on definitions of attitude.
  • There are no generally agreed methods for measuring attitude.

The technical research literature on attitude is vast and this is necessarily a brief summary. But it reflects the major findings from over 70 years of study.

The only finding that might encourage anyone to persist with the idea of attitude is of no use in communication, as it shows that so-called ‘attitude change’ follows behaviour, rather than the other way round. On some occasions, after people’s behaviour has been changed, their responses to survey questions also change. For example, in the USA, following the introduction of mixed-race working groups that had been previously segregated, workers gave less prejudiced responses to survey questions. But does a changed response to a survey question signal a change in something called ‘attitude’? Unfortunately, nobody knows.

Attitude change in communication campaigns

While psychologists may wish to argue for years to come about these findings and their implications for social psychology, public relations practitioners are faced with a much more immediate and pragmatic problem. Should they continue to use ‘attitude’ measurement and change as part of the objectives in communication campaigns?

Communication campaign managers have used the idea of attitude in a relatively simple way, arguing that their particular campaign will affect public ‘attitudes’ which in turn will lead to changes in public behaviour. This is not good science, and is unlikely to lead to predictable campaign outcomes.

Achievable campaign objectives

Predictable campaign outcomes are, however, still possible. In our own research we have found no need to resort to concepts of attitude or to use attitudinal measures, yet we still achieve worthwhile and measurable results.

From our point of view as communication researchers, ‘attitude’ looks increasingly like the phlogiston of the mind: an interesting idea that has proved untenable and irrelevant.

We would therefore advise against campaign strategies that include attitude change as an objective, and against studies that try to measure attitude. This will enable you to run campaigns with realistic objectives. Moreover, you will be able to save money, since you do not need to run expensive surveys to collect flimsy data in support of dubious arguments.

Based on our experience, we suggest four basic questions around which to construct campaign objectives:

  • what do people know?
  • what do they understand?
  • what can they do?
  • what do they do?

These questions lend themselves much more readily to observation and answers. You can use campaigns to change what people know and understand. You might even be able to use them to show people what they can do. But if your ultimate objective is to change what people do, then you are more likely to succeed if you change the environment in which they act, and the controls over their actions. In the end you may have to get a change in legislation to achieve changes in behaviour. And, if—as the research suggests—changes in behaviour do on occasion lead to changes in responses to attitude surveys, you can still pander to the belief in the alchemy of attitude, and measure ‘attitude change’, just as the early chemists ‘measured’ phlogiston. end