Construction and validation of habit measures for fruit and vegetable consumption

Christopher Rompotis

Research output: ThesisDoctoral Thesis

296 Downloads (Pure)


Adequate fruit and vegetable consumption is a protective factor against chronic illness, and increasing fruit and vegetable consumption is a public health priority. While public health interventions have modestly improved population fruit and vegetable consumption, only 48.5% and 8.2% of Australian adults currently consume the recommended daily servings of fruit and vegetables, respectively. Social, environmental and economic predictors have been identified as factors that affect fruit and vegetable consumption. However, these predictors are difficult and expensive to modify. Alternatively, psychological predictors may be modifiable and cost-effective mechanisms for increasing fruit and vegetable consumption. Dual-processing frameworks, which models behaviour using reasoned processes (e.g. intentions, self-efficacy and attitudes) alongside automatic and habitual processes, may provide a useful approach in predicting fruit and vegetable consumption. However, previous research has been limited by its focus on reasoned processes at the expense of habitual processes. Consequently, measures of habitual processes require further development. Furthermore, previous research has examined the relationships between psychological predictors and combined fruit and vegetable intake, despite indications that these relationships may differ between the two behaviours. Lastly, there has been a lack of experimental research to support the use of habit-based interventions for improving fruit and vegetable consumption in adults. The present thesis addresses these limitations by exploring the relationships between multi-process habit measures (automaticity, patterned response, stimulus-response bonds and negative consequences for non-performance), measures of cognitive processes (e.g. Theory of Planned Behavior), and fruit and vegetable consumption. Furthermore, the effectiveness of an informational intervention based on a multi-process habit framework is assessed.

Experimental chapters in this thesis are presented as a series of papers. Four studies are included which examine measures of habit strength, fruit and vegetable consumption and cognitive processes in Western Australian adults (N = 619). In studies one and two, psychometric support is given for a multi-process habit measure of combined fruit and vegetable consumption. Specifically, automaticity of consumption, response patterning (routinisation) and negative psychological consequences for non-consumption are positively related to fruit and vegetable consumption, and stronger in those eating at least five serves of fruit and vegetables per day than those who do not yet consume at least five serves. Study three indicated that the multi-process habit framework was supported for examining fruit consumption. Furthermore, the multi-process habit measures accounted for additional variance in fruit and vegetable consumption above that of automaticity alone. However, when reasoned processes (i.e. Theory of Planned Behavior) were incorporated into the model, habit processes did not significantly predict fruit or vegetable consumption. Study four demonstrated that a habit-based informational intervention resulted in a greater change in fruit consumption than messages based on meal preparation strategies or healthy eating advice. However, habit-based messages were equally effective in improving vegetable consumption when compared to meal preparation strategies and healthy eating advice.

This thesis demonstrates the importance of incorporating a multi-process habit framework alongside traditional measures of cognitive predictors of fruit and vegetable consumption. Additionally, a multi-process habit framework (automaticity, routinisation and negative psychological consequences for non-performance) may account for additional variance in fruit consumption and vegetable consumption than measures of automaticity alone. Furthermore, habitual and cognitive processes were found to predict fruit consumption and vegetable consumption differentially. Future research and promotion campaigns may benefit from targeting the behaviours separately.
Original languageEnglish
QualificationDoctor of Philosophy
  • Grove, Bob, Supervisor
  • Byrne, Susan, Supervisor
Publication statusUnpublished - 2015


Dive into the research topics of 'Construction and validation of habit measures for fruit and vegetable consumption'. Together they form a unique fingerprint.

Cite this