ABSTRACT
In the previous chapter, you learned that attitudes do indeed serve as predictors of behaviours. However, you will remember that attitudes are not always reliable predictors, but that measurement aspects may moderate the attitude-behaviour correlation. If there was a mismatch between attitude and behaviour measure – for example, with respect to specificity – correla - tions may only be weak. As we showed in the previous chapter, much of the failure to predict behaviour from attitudes (alone) can be explained by weak measure correspondence. It is noteworthy that the correspondence principle, more broadly conceived, allows for many predictions beyond the measure itself. More generally speaking, attitudes and behaviour are more closely related if the aspects of the attitude that are highly accessible at the time of attitude measurement are also accessible at the time the behaviour is performed (Shavitt & Fazio, 1991; Tesser & Shaffer, 1990; Wilson & Dunn, 1986; Wilson et al., 1989). To test this hypothesis with respect to attitude functions, Shavitt and Fazio (1991) examined students’ attitudes towards two brands of soft drink: 7-Up lemonade and Perrier mineral water. Pilot testing had shown that attitudes towards each drink typically served distinct functions. Perrier was liked mainly because of the trendy image it imparts (social image function), whereas 7-Up was liked foremost for its taste (utilitarian function; see Chapter 1). In their experiment, Shavitt and Fazio primed either image or taste before participants were asked to evaluate each drink: specifically, some participants judged the social impression that each of 20 behaviours would create, whereas others rated 20 food items for their taste. Later, all participants reported their attitudes towards both 7-Up and Perrier along with their intentions to buy each of these drinks. Shavitt and Fazio hypothesised that attitudes would more strongly predict behav - ioural intention if the primed attitude function matched rather than mis - matched the function that is normally associated with the attitude object. As shown in Figure 13.1, the data supported this prediction. Students’ intentions to purchase Perrier were more highly correlated with attitudes towards Perrier when they had thought about social impression rather than taste just prior to attitude measurement. By contrast, students’
intentions to buy 7-Up were more highly correlated with their attitudes towards 7-Up when they had previously thought about taste rather than social impression. Therefore, marketers should not only consider how to influence an attitude, but should also take the conditions at the point of sale into account. Creating a positive attitude which serves a symbolic function may not affect purchase if in the purchase situation more utilitarian attitudes prevail. To profit from the positive attitude, it might then be worthwhile to further activate thoughts about social situations.
