Selection and quantification of typical food portions by young adults.

This study was designed to assess (a) the amount of food young adults selected as an average portion and (b) the accuracy with which they could, without the assistance of food models or probing by an interviewer, describe the amount in quantifiable terms. One hundred and forty-seven adults between 18 and 30 years of age selected average portions of either breakfast or lunch items and then described the amount. Selected amounts of butter on toast, sugar on cereal, milk as a beverage, and tossed salad corresponded closely with generally accepted standard portion sizes. For dry cereal, orange juice, tuna salad, and fruit salad, selected portions were above accepted standards, but the portion for salad dressing was below the standard. Men almost consistently selected larger portions than did women. The ability to describe the amount chosen without the aid of measuring devices was poor, with from 8% to 68% of respondents being able to estimate individual items within +/- 25% of actual amount; from 0 to 67% overestimated portion size by more than 51%; and from 0 to 25% underestimated by more than 51%. The results suggest a need to provide respondents with help in estimating portion sizes or to develop alternative methods of assessing dietary intake that do not assume an ability to describe portion size accurately.