The effects of behaviour on attitudes

The effects of behaviour on attitudes

The research issue

Before this experiment was conducted, most researchers were primarily interested in the effects of attitudes on behaviour.In contrast, Festinger and Carlsmith (1959) set out to show that our behaviour can occasionally be awkwardly inconsistent with our true attitude and, to resolve this uncomfortable inconsistency, we may change our attitude to match the behaviour. The experiment consisted of three stages. In the first stage, the experimenter attempted to make participants dislike a series of boring tasks. Participants were falsely told that they were taking part in a study of ‘measures of performance’.They were asked to put 12 spools on a tray, empty the tray, and refill it. Participants repeated this task for half an hour, using one hand, while the experimenter pretended to record their performance. Next, participants were asked to use one hand to turn 48 square pegs on a board (a quarter turn one way, then the other way) for half an hour, while the experimenter continued to monitor their performance. Presumably, participants came to hate these dull tasks. In the second stage, the experimenter asked the participants to tell a new participant that the tasks were interesting and enjoyable. The experimenter justified this request by stating that he was comparing the performance of participants who had been told nothing about the task with the performance of participants who had been gi ven specific, positive expectations.The experimenter indicated that his colleague usually gave specific (positive) information to participants, but that this colleague had not arrived yet.The experimenter then asked whether the participant could temporarily fill in and be ‘on call’ for future elements of the study. Virtually all of the participants agreed to this request. The participants then attempted to persuade the next participant (who was actually a confederate of the experimenter) that the tasks were interesting, fun, enjoyable, intriguing and exciting.In the third stage, participants were asked to meet an interviewer to answer questions about the previous tasks (e.g. turning the pegs). One of the questions was about the extent to which participants enjoyed the tasks. Festinger and Carlsmith expected that participants’ intervening behaviour would cause them to like the tasks to a greater extent only when they believed they had been given little external incentive for engaging in the behaviour. If the behaviour was performed with little reward, participants should feel a need to justify the behaviour to themselves. To do this, they should change their attitude to support the behaviour. In other words, participants should come to believe that they actually liked the tasks that they had undertaken during the intervening period. To test this reasoning, the experiment included a crucial manipulation: participants were offered either $20 (a lot of money in the 1950s!) or $1 to describe the dull tasks favourably to the other ‘participant’.
Results and implications
As shown in figure 17.5, the results support this prediction. After the experimental manipulation, participants were more favourably disposed towards the tasks if they had been offered $1 than if they had been offered $20. In addition, participants who were offered $1 were more favourably disposed than those who were not asked to say anything about the tasks (control condition). Overall, these results support cognitive dissonance theory by showing that people can alter their attitudes to justify their past behaviour. Since this experiment, abundant research has shown that this attitude change helps to reduce an unpleasant arousal that people experience after performing the attitude-incongruent behaviour, while also finding some limitations to this effect (see Cooper & Fazio, 1984).

Festinger, L., & Carlsmith, J.M., 1959, ‘Cognitive consequences of forced compliance’, Journal of Abnormal and Social Psychology, 58, 203–10.

HOW ATTITUDES INFLUENCE BEHAVIOUR

HOW DO ATTITUDES INFLUENCE BEHAVIOUR?

Ever since the beginning of attitude research, investigators have puzzled over the relation between attitudes and behaviour. Why do people sometimes say they like something and then act as if they do not? Are these instances much less frequent than instances where the attitude and behaviour match perfectly?
Measuring the attitude–behaviour link Researchers were intrigued by the results of some early research that revealed very weak relations between attitudes and behaviour.In one study (LaPiere, 1934), a researcher and a young Chinese couple travelled around the Western portion of the US, visiting 250 restaurants, inns and hotels. Despite widespread American prejudice against Chinese people at that time, the researcher and his visitors were refused service at only one of the establishments. Yet, when he later wrote to these establishments requesting permission to visit with ‘a young Chinese gentleman and his wife’, 92 per cent refused permission! These refusals are often interpreted as indicators of negative attitudes towards Chinese people. Viewed this way, they provide someof the earliest evidence that people’s behaviours (in this case,accepting the Chinese couple) can fail to match their attitudes towards the behaviour (i.e. their desire to refuse permission).This raised some doubts about the ability of attitudes to predict behaviours.
There were many methodological limitations to LaPiere’s study, however (Campbell, 1963). For example: the attitude and behaviour were measured at different times and locations; the attitude measure itself was, at best, indirect (LaPiere did not ask the restaurant owners to complete an attitude scale); the young couple may have looked more pleasant than the proprietors had imagined; the proprietors may have followed the norm of hospitality to guests once they entered the restaurant; and the situation in which behaviour was measured may simply have made it too difficult for most proprietors to refuse the Chinese couple, because of the embarrassing scene that might ensue.Subsequent studies used more stringent procedures . Using a correlational technique, these studies tested whether people with positive attitudes towards a particular object exhibit more favourable behaviour towards the object than do people with negative attitudes towards the object. Even so, until 1962, researchers still found only weak relations between attitudes and behaviour.
The consistent failure to find strong attitude–behaviour correlations led researchers to search for explanations. Fishbein and Ajzen (1975) pointed out that past research often failed to measure a behaviour that directly corresponded to the attitude being measured. For example, suppose we measure the relation between (a) attitudes towards protecting the environment and (b) using a recycling facility in a particular week. Even if someone is a strong environmentalist, there are many reasons why they might fail to recycle in a particular week (lack of a nearby facility, lack of time to sort recyclables, and so on). The problem is that the measured behaviour (recycling in a particular week) is very specific, whereas the attitude object (protecting the environment) is much more general. To better measure ‘general’ behaviour, Fishbein and Ajzen (1975) proposed the multipleact criterion, which involves measuring a large number of behaviours that are relevant to the general attitude being studied. For example, to measure pro-environment behaviour, we could measure numerous proenvironment behaviours, including recycling across several weeks, willingness to sign pro-environment petitions and tendency to pick up litter. This would give us a more precise and reliable measure of behaviour. Weigel and Newman (1976) did just this and found much stronger attitude–behaviour relations by takingan average measure of all of the behaviours, rather than any single behaviour