The aim of this study was to assess the validity of the RT3 accelerometer, and its inbuilt algorithm, in measuring inactivity, walking and running in children. Twenty children, aged 7-12 years, participated in the study. The RT3 was compared to physiological energy expenditure obtained via a wireless portable ergospirometric system. Data analysis was performed using Bland and Altman plots and Pearson product moment correlation coefficients. There were no significant differences between the methods for each activity. Bland and Altman 95% limits of agreement between the two measures in kcalmin(-1) for each of the activities were as follows: inactivity (-0.058, 0.47), walking at 3kmh(-1) (-1.22, 0.83) brisk walking at 6kmh(-1) (-2.74, 0.54), brisk walking at 6kmh(-1) on an incline of a 10% gradient (-1.69, 2.11), and jogging at 9kmh(-1) (-3.67, 1.24). Energy expenditure via the RT3 correlated significantly with that obtained by indirect calorimetry for each activity independently (r=0.56-0.84, all P<0.01). The RT3 provided a valid estimate of inactivity, walking and running and would thus appear appropriate for the objective measurement of physical activity levels.