This study compared the accuracy with which respondents reported food intake using different portion-size estimation aids (PSEAs) and determined whether the interview technique affected the accuracy of reporting.

Participants chose preweighed foods from a buffet line, leftovers were weighed, and actual amounts eaten were calculated. The next day during dietary interviews, participants estimated amounts eaten using one of four procedures, including telephone or in-person interviews, two-dimensional or three-dimensional PSEAs, and guiding or not guiding participants to certain aids.

A total of 120 persons, ages 18 to 65 years, representing both genders were recruited through advertisements, local churches, and universities.

Analysis of variance and least significant differences were calculated for mean percentage estimation error. Frequencies and Pearson correlation coefficients were used to determine percentage misestimations and relationships between estimations and perceived healthfulness of the food and confidence in the estimate made.

In general, accuracy of reporting food amounts was not significantly different for the different types of aids or for the interview technique. Participants frequently overestimated the intake by more than 20% for solids (four overestimations and zero underestimations) and liquids (13 overestimations and one underestimation), and both over- and underestimated amorphous (mounding) foods by more than 20% (six underestimations and seven overestimations). Misestimation was unrelated to perceived healthfulness of the food or respondent confidence in the estimate.

In this study, portion-size data, based on PSEAs used in dietary recalls for estimating amounts eaten, were useful for measuring dietary intake for population-based studies. Individual respondent data, however, may be inaccurate and should be supplemented with additional information to improve assessment of individual nutritional status. Techniques such as telephone interviewing and guiding respondents to certain aids seem useful for reducing the burden of dietary assessment.