Decoding the misperception: Exploring measurement error in self-rated assessments of diet quality

Abstract

How much should we trust self-rated diet quality? Measures based on self-perception are often used in diet evaluation, but they can sometimes be subject to diverse biases. Using a unique setting provided by the nationally representative FoodAPS data where questions about diet quality are asked twice with intermediate reminders in between, we first show that the structure of survey questions can help identify inconsistency in the respondents’ self-rated diet quality. Findings reveal that respondents tend to deliver responses that align with their earlier responses. We then validate the respondents’ self-rated diet quality against their household’s healthy eating index, an objective measure of diet quality based on actual food acquisitions. We find no statistically significant association between self-rated diet quality and healthy eating index for respondents who display inconsistency in self-assessing their diet quality. Conversely, that relationship is significant for respondents with no such inconsistency. Our findings suggest an effective question design measuring diet quality, that can be generalized to the design of health survey questions.