The purpose of this study was to determine how well a group of ten men and eight women who attended one of four dinners could estimate measured and weighed portions of ten test foods by a 24-hour dietary recall over the telephone, with and without a food model book.

A paired t-test indicated that there was no statistically significant difference between the 24-hour dietary recalls estimated by the total group with and without the food model book. However, the standard deviation for the dietary recall using the food model book was less than half that of the recall from memory alone. While neither method is wholly reliable because of the large standard deviation, the food model book method appears to be more reliable for estimating total food intake than the home measures method which involves recall from memory alone. When the data were analyzed separately by sex, the differences between the results obtained with the use of the food model book method and of the home measures method were not significant for either women or men.

The mean estimated volumes were found by t-tests to be significantly different from the actual volumes served for six out of ten foods when using the food model book method, and for two out of ten foods when using the home measures method. These results raised some questions as to the validity of these methods for estimating food intake and indicated that the group was giving biased estimates of Estimated volumes of four foods recalled by the food model book method and eight recalled by the home certain foods. measures method were found not to differ significantly from the actual served volumes.

Standard deviations indicated considerable variation in the ability of the subjects to estimate the ten test foods. This was especially true for the home measures method where the standard deviations for all foods were consistently larger than when subjects used the food model book method.

For both methods, there was a much greater tendency to underestimate than to overestimate the volumes of food items served, and the underestimates were larger in percentage of mean absolute differences.

When the mean estimated food intake was compared to the actual food intake by t-tests, women and men estimated alike. There was no difference by sex in their ability to estimate food intake during a 24-hour recall by using either the food model book method or the home measures method but neither was wholly reliable or accurate. For both women and men, the difference between estimates by the food model book method and the actual volume of food served was significant. In the home measures method, the differences were not significant for either women or men. As the standard error for the food model book was smaller. that method proved to be more sensitive, i.e., smaller absolute differences were significant. This may account for the significant difference of the mean estimated food intake from the actual food intake for the food model book and the nonsignificance for the home measures method.

When the mean estimated food intakes for women and men were compared by a t-test which pooled the standard deviations, there was no significant difference in the performance of the two sexes.

The ten foods tested were classified into a geometric or nongeometric category based on shape. There was no significant difference between the mean scores of geometric and nongeometric foods, between women and men, or between the two methods.

As geometric foods have a definite shape, it was expected that they would be easier to estimate than nongeometric foods which have an ill-defined shape. However, such geometric foods as the cookie and the vegetarian loaf were underestimated considerably. This may have occurred because of the difficulty in estimating thickness, since a fraction of an inch can make a considerable difference in the estimated volume.

Since the use of the food model book has been shown to decrease the standard errors of the estimates for all the food items tested, it appears likely that this method in conjunction with an educational program prior to a 24-hour dietary recall could increase the subject's awareness of the types and the amounts of food eaten. Thus, it might be possible to improve further the reliability of food recall.


Graduate School

First Advisor

Irma B. Vyhmeister

Second Advisor

Ezbon Jen

Third Advisor

Helen H. Register

Fourth Advisor

Grenith J. Zimmerman

Degree Name

Master of Science (MS)

Degree Level


Year Degree Awarded


Date (Title Page)




Library of Congress/MESH Subject Headings

Nutrition Surveys



Page Count

x; 109

Digital Format


Digital Publisher

Loma Linda University Libraries

Usage Rights

This title appears here courtesy of the author, who has granted Loma Linda University a limited, non-exclusive right to make this publication available to the public. The author retains all other copyrights.


Loma Linda University Electronic Theses and Dissertations

Collection Website



Loma Linda University. Del E. Webb Memorial Library. University Archives