The accuracy of children's self-reports of diet: Family Health Project

Abstract

This study was designed to test the accuracy of four different methods for enabling children in the third to sixth grades to record their frequency of consumption of foods high in any of four targeted nutrients. The forms varied in two ways: recording the whole day or a segment of the day (morning, afternoon, or evening) or use or non-use of pictures of the food items. The accuracy of the children's recording of food consumption was validated by observation of their behavior for 2 continuous 12-hour days. Twenty-four children in the third to sixth grades were observed for each of the 2 days. An 82.9% agreement was obtained between the child's self-reported food frequency and the observer's record of the child's consumption. Ethnicity slightly affected the accuracy of form completion, while gender and grade level of the children did not. The results of this study validate the accuracy with which children record diet using a food frequency of consumption method.