Accuracy of fourth- and fifth-grade students' food records compared with school-lunch observations.

Fourth- and fifth-grade students' food records completed by using a labor-intensive approach (with daily monitoring) or a less labor-intensive approach (with weekly monitoring) were compared with school-lunch observations. Records were completed at three elementary schools for 14-23 d by students in 26 classes. Record pages were printed forms with areas for recording items consumed, number of servings, and locations of meals. Students were randomly selected for lunch observation. Kappa coefficients and percent of agreement for interobserver reliability for both approaches combined ranged from 0.43 to 0.91 and 72.6% to 95.5%, respectively, on 24 observations. Pearson correlations for 117 record/observation matches were significant for eight and three of nine meal items for the daily and weekly monitoring approaches, respectively. Thus, students were reasonably accurate in keeping records and more accurate with the more labor-intensive approach. Researchers need to select an approach based on the degree of accuracy desired and available finances.