In 1988, the Yale Physical Activity Survey (YPAS) was designed and then administered to healthy older populations of volunteers (aged 60-86) to establish its 2-wk repeatability and relative validity. Among the 76 volunteers in the repeatability substudy, correlation coefficients between the two administrations of the survey for the eight YPAS summary indices ranged from 0.42 (P = 0.0002) to 0.65 (P = 0.0001). Among the 25 subjects in the validation substudy, weekly energy expenditure (r = -0.47; P = 0.01) and daily hours spent sitting (r = 0.53; P = 0.01) correlated with resting diastolic blood pressure, while the YPAS activity dimensions summary index (composed of questions on vigorous activity, leisurely walking, moving, sitting, and standing) correlated positively with estimated VO2max (r = 0.58; P = 0.004) and inversely with percent body fat (r = -0.43; P = 0.03). The YPAS index of vigorous activity also correlated positively with estimated VO2max (r = 0.60; P = 0.003) and the moving index correlated marginally with body mass index (r = -0.37; P = 0.06). We conclude that the YPAS demonstrates adequate repeatability, and some validity by correlating with several physiologic variables reflecting habitual physical activity. The value of the YPAS, however, in accurately assessing low intensity activity remains to be established.