Structure and content of affective associations with health behaviours: Is the behaviour 'good OR bad' or 'good AND bad'

Objective: The feelings and emotions individuals associate with health-related behaviours influence engagement in those behaviours. However, the structure and the content of these affective associations have not been examined. The studies reported here examined competing hypotheses about the structure (unidimensional or bidimensional) and content (generalised affect or specific emotions) of affective associations with two health-related behaviours: physical activity and fruit/vegetable consumption. Design: For each behaviour, participants (fruit and vegetable consumption n = 149; physical activity n = 199) completed an assessment of the association of 40 positive and 51 negative affect concepts with the behaviour. Main outcome measures: Ratings of affective associations with each behaviour. Results: Confirmatory factor analyses comparing unidimensional and bidimensional affect structure models showed that the structure of individuals' affective associations was bidimensional for both behaviours - positive and negative affective associations were shown to be separate and distinct constructs. Exploratory factor analyses supported a model of affective associations as generalised affect for both behaviours. Conclusion: Affective associations with both physical activity and with fruit/vegetable consumption consist of separate positive and negative dimensions of generalised affect. These findings lead to recommendations for research and intervention development based on the implications for how affective associations might operate to influence behavioural decision-making.