Confidence Nutrition Study 5

Previous research has demonstrated robust associations between confidence in one’s knowledge and a range of decision-making outcomes, including self-efficacy and information preferences. In particular, in previous work we conducted, we found that confidence (operationalized as perceived knowledge) was positively associated with interest in an app add-on that provided complex nutritional information, but was negatively associated with interest in an app add-on that provided simplifying information. However, because confidence was examined without controlling for objective knowledge, these effects could reflect true knowledge rather than confidence beyond knowledge. To address this limitation, the present project introduces the Nutrition Knowledge and Confidence Assessment (N-KCA), which provides a set of 40 items asking about people’s knowledge and confidence in their knowledge. This allows us to isolate unjustified confidence—that is, confidence not accounted for by objective nutrition knowledge. The primary purpose of this study is to replicate and extend our previously observed relationships between confidence and nutrition information search behavior. Our earlier work was the first to document both positive and negative associations between confidence and preferences for different types of nutrition information tools within the same study. In the present study, we will replicate these findings both with and without controlling for objective knowledge, allowing us to determine whether these previously observed relationships remain robust when knowledge is measured and statistically controlled. Confidence will be operationalized in two ways. First, we will measure confidence from the N-KCA. Second, we will use the same perceived knowledge scale used in the previous work. In addition, we will examine associations between confidence and self-efficacy, and between self-efficacy and preferences for information-rich versus information-light nutrition tools. As a secondary aim, we will examine the structure of self-efficacy in greater detail. In our previous study, self-efficacy was primarily treated as a unidimensional construct, and it showed relationships with app add-on preferences similar to those observed for perceived knowledge. However, there was also preliminary evidence suggesting that self-efficacy may consist of two distinct components: (1) self-efficacy in one’s ability to use existing knowledge to make nutrition-related decisions, and (2) self-efficacy in one’s ability to use existing knowledge to interpret additional nutrition information. These components appeared to show differential associations with app add-on preferences. The present study will evaluate whether self-efficacy is better characterized as a single construct or as a multidimensional construct with distinct predictive patterns. All primary analyses will be conducted both with and without controlling for objective knowledge. In all models, interest in nutrition apps will be included as a covariate. In addition, app evaluations will be analyzed separately and comparatively for ratings made for oneself and for a typical person, allowing us to examine whether these relationships are specific to self-relevant judgments.