Self-reported weight and height are used to compute ponderal indices in studies where it is impractical to obtain objectively measured data. Little is known about the associations between self-reported and measured weight and height in older adults. Participants in congregate and home-delivered meal programs in a rural Wisconsin county recorded their height and weight on a questionnaire as part of a study on nutritional risk and quality of life. Objective measurements of height and weight were made on the same day. Data for 131 subjects, ranging in age from 62-92 years, were analyzed to evaluate associations between self-reported and measured weights and heights. Correlation coefficients between self-reported and measured values suggested that self-reported weight was somewhat more valid than self-reported height in this population. However, self-reported weight was misreported by 10 pounds or more in 20% of the sample. More data are needed to evaluate the utility of self-reported data for weight and height in groups of older people. Judging from this study, self-reported height and weight data, if used at all for older adults, should only be used as a continuous variable rather than for classifying subjects into relative weight categories.