Fast-food consumption: application and extension of the theory of planned behaviour to incorporate affective responses and implicit associations.

Abstract

The consumption of energy-dense fast foods has been implicated as a causal factor in the development of obesity. The development of strategies to modify food choice behaviour requires an understanding of both the behaviour and the influencing factors (in particular, beliefs, attitudes, and social influences). The Theory of Planned Behaviour (TPB; Ajzen, 1988) is one of the most widely used and accepted models of the intention-behaviour relationship within the health literature (Bagozzi, Wong, Abe, & Bergami, 2000; Conner & Armitage, 1998; Sheppard, Hartwick, & Warshaw, 1988; Sutton, 1998). This thesis spans three studies, applying and extending the TPB to examine factors influencing fast-food consumption in an Australian population. The first study is a qualitative analysis examining components of the belief stage of the TPB, including behavioural outcome, normative, and control beliefs in relation to frequent fast-food consumption. Aside from details of these beliefs, which were used to develop a quantitative measure for application in the second study, the key outcomes of this study were a definition of fast food which was applied throughout each of the studies, clarification of how fast food fits within Australian lifestyles, and generation of additional information regarding other possible influences on fast-food consumption beyond those addressed by the TPB. The additional variables included for examination were affective responses to fast food, individual differences in sensitivity to reward, and the extent to which both consideration of the future consequences of frequent fast-food consumption and fear of being negatively evaluated are influential on consumption rates. The second, quantitative study applied both the TPB-based instrument developed from findings of the first study as well as the four additional measures with the aim of improving the explanatory ability of the TPB in terms of fast-food consumption. Structural equation modelling was used and although the model performed well overall, explaining up to 50% of the variance in both intention and behaviour, there were some variables that did not perform strongly. In particular, the measure of affective responses captured with a semantic differential scale was not significantly predictive, and the third study was designed and conducted to test the performance of a more sophisticated measure in the hope that implicit, in addition to explicit, responses would provide greater explanatory value. The Implicit Association Test (IAT; Greenwald, McGhee, & Schwartz, 1998; Greenwald, Nosek, & Banaji, 2003) is said to provide some insight into more implicit attitudes by measuring the degree to which an individual associates conceptual categories through response latencies (or timed responses) whilst opportunities for introspection are minimised. Some have also suggested that the IAT may contribute well when it is combined with other measures, particularly in the prediction of behaviour, choice, or judgement (Brunel, Tietje, & Greenwald, 2004). Therefore, the third study combined measures based on the TPB with the IAT to investigate the nature of both implicit and explicit beliefs underlying attitudes towards fast foods. Overall, the associations captured by the IAT did not help to explain a great deal of variance in fast-food consumption, although there were differences in associations according to stimulus type with arousal-related stimuli generating stronger results than valence-related stimuli. Differences in associations with fast food were also found across groups divided according to Body Mass Index, with overweight people demonstrating more positive associations than either normal weight or obese people. A summary discusses the potential application of the findings in the development and implementation of future intervention and obesity prevention strategies as well as directions for future research.