De Bruijn, G. J., Out, K., & Rhodes, R. E. (2014). Testing the effects of message framing, kernel state, and exercise guideline adherence on exercise intentions and resolve. British Journal of Health Psychology, 19(4), 871-885. doi: 10.1111/bjhp.12086

Abstract:
Objectives: To study the effects of framed messages on exercise intention and resolve.
Design: Two (type of frame: gain or loss) × 2 (type of kernel state: desirable or undesirable outcome) post-test study.
Methods:  Participants were recruited online and questioned about their previous exercise behaviour and their exercise risk perception. After this, they were randomly allocated to one of four messages that were different in terms of positive or negative outcomes (type of frame) and in terms of attained or avoided outcomes (type of kernel state). After reading the message, participants indicated their intention and resolve to engage in sufficient exercise.
Results: No effects were found for intention. For resolve, there was a significant interaction between type of frame, type of kernel state, and exercise adherence. Those who did not adhere to the exercise guideline and read the loss-framed message with attained outcomes reported significantly higher resolve than all other participants.
Conclusions: This study indicates the relevance of including attained outcomes in message framing exercise interventions as well as a focus on exercise resolve.
Statement of Contribution: What is already known on this subject? Message framing is commonly used to increase exercise intentions and behaviour. Meta-analyses do not provide consistent support for this theory. Very little attention has been paid to resolve and message factors on framing effects. What does this study add? Framed messages have an effect on exercise resolve, but not on intention. Loss-framed messages with attained outcomes are most persuasive for those who do not adhere to exercise guidelines. Exercise framing studies should include behavioural resolve next to intention.

Link: click here