Improved methods for quantitative self-reports of total physical activity in epidemiological studies are needed. We evaluated randomly selected individuals' ability to integrate their perception of physical activity over time to produce an estimate of the "usual" level, using a novel instrument for self-quantification of energy expenditure. A population-based sample of 418 Swedish men and women, age 20-59, completed a questionnaire containing the new instrument. For validation, three 24 hour recalls by phone served as gold standard. Reproducibility was assessed through administering the instrument another three times. The validation involved 133 subjects and another 160 completed the reproducibility evaluation. Pearson correlation between usual daily energy expenditure measured by the instrument and the mean of the 24 hour recalls was 0.73. After subdividing the self-reported daily energy expenditure and the mean of the 24 hour recalls into quintiles, 83.5% of the participants remained in the same quintile, or one quintile apart. There was a tendency towards overestimation of usual daily physical activity. This was significantly associated with low education. Reproducibility showed an intraclass correlation of 0.55. Although integrated reports of usual daily energy expenditure over longer periods seem to be afflicted with a tendency of overestimation, total energy expenditure can be estimated with reasonable validity and reproducibility using our instrument.