Aim of this study was to assess the validity of a short self-administered physical activity questionnaire, intended to measure past year total daily physical activity, by comparison with activity records and accelerometers. Over a 1-year period, data from a questionnaire, 7-day activity records and accelerometers were obtained from a subset of 116 women between the ages of 56 and 75 years from the population-based Swedish Mammography Cohort. We estimated concordance correlations as measure of validity, deattenuated for intraindividual variation in the reference method. Deattenuated concordance correlations comparing total daily activity measured by the questionnaire with the accelerometers and the records were 0.38 (95% CI: 0.22-0.54) and 0.64 (95% CI: 0.45-0.83), respectively. Validity of leisure-time activity (walking/bicycling and exercise) and inactivity (watching TV/reading) estimates comparing the records with the questionnaire were 0.42 (95% CI: 0.22-0.62) and 0.52 (95% CI: 0.36-0.69), respectively. These data indicate that the average past year total physical activity, leisure-time activity and inactivity can be estimated with a reasonable validity using our short self-administered questionnaire.