The accurate measurement of physical activity is fraught with problems in adults, but more especially in children because they have more complex and multi-dimensional activity patterns. In addition, the results of different studies are often difficult to interpret and compare, because of the diversity of methodological approaches, differences in data analysis and reporting, and the adoption of varying definitions of what constitutes an appropriate level of activity. Furthermore, inactivity is seldom quantified directly. Although there exists an extensive literature documenting the health benefits of regular physical activity in adults, activity-health relationships in children are not clear-cut. Current recommendations reinforce the concept of health-related activity, accumulating 30 min moderate-intensity exercise on at least 5 d/week (adults) and 1 h moderate-intensity exercise/d (children). Evidence suggests a high prevalence of inactivity in adults, but whether or not inactivity is increasing cannot be assessed currently. Similarly, no definite conclusions are justified about either the levels of physical activity of children, or whether these are sufficient to maintain and promote health. Data to support the belief that activity levels in childhood track into adulthood are weak. Inactivity is associated with an increased risk of weight gain and obesity, but causality remains to be established. In children there is strong evidence to demonstrate a dose-response relationship between the prevalence and incidence of obesity and time spent viewing television. Future research should focus on refining methodology for physical activity assessment to make it more sensitive to the different dimensions and contexts of activity in different age-groups.