Comparison of Three Visual Methods for Estimation of Plate Waste

Abstract

LEARNING OUTCOME: To compare three visual methods for estimation of plate waste. Food consumption is used to evaluate the acceptability of foods, adequacy of intake, and effectiveness of nutrition education interventions. Plate waste frequently is used to assess food consumption in group feeding programs. Estimates of plate waste have traditionally been calculated by the method of weighing each item individually which is inconvenient and time-intensive. Several studies have compared visual estimates and weighed waste. Others have reported using videotaping to identify amounts consumed. The objective of this study was to compare three methods of estimating plate waste. The three visual methods compared were video (V), computer screen (CS) and computer print (CP). Five menus representative of typical school lunches were planned. For each menu, six control trays and twelve variation trays were prepared. Portions of menu items were removed according to a randomly assigned plan to represent consumption of zero to 100 percent of each item. Initial weights and weights of portions removed were recorded for each item. A video camera on a tripod was positioned at a 45° angle above the counter for taping. A cube of known dimension and a 12-inch ruler were positioned adjacent to the tray. Each tray was videotaped for 15 seconds. The videotape was reviewed on a 22-inch monitor. A video capture device was used to convert the video of each tray to digital images for CS and CP viewing. A registered dietitian trained in visually estimating plate waste viewed the trays and evaluated portions remaining using the Comstock scale of 1=25% of an item remaining, 2=50% of an item remaining, 3=75% of an item remaining, 4=95% of an item remaining, and 5=100% of an item remaining. Three hundred and thirty-six servings of 24 menu items were evaluated. Data was statistically analyzed using the Statistical Package for the Social Sciences for personal computers. Pearson correlation coefficients were used to compare individual values obtained by the three methods of quantitating foods with actual portion weights. Correlation coefficients ranged from 0.87 to 0.94 (p≤0.01) indicating strong correlation. The research results reveal that there were no significant differences among the visual methods.