This research examined two concerns. The first was theoretical: A study of the nature of self-efficacy and its relationship to predictors, including demographics as well as outcome variables relating to diet, was carried out. A second concern was the use of two dimensions of self-efficacy to produce a segmentation scheme that could be used to improve the efficiency of public health campaigns, in this case a community-based information campaign on dietary change. The data were collected via telephone interviews prior to the intervention. Hypotheses concerning relations of several variables to our two dimensions of self-efficacy found some support and some nonsupport, providing valuable evidence to detail the importance of self-efficacy in the health belief model and in public health interventions. The self-efficacy dimensions were used to design messages used in the community intervention. A short discussion of this application is provided.