To examine in the context of a school day whether a pedometer, compared to an accelerometer, was an effective tool to identify children with low physical activity levels. Additionally, to characterise within the school day, patterns of variation in level and timing of activity of the most active and least active children.Observational study to compare two methods (accelerometer and pedometer) of identifying less active children and intensity of activity.During school hours, for three consecutive days, children (n=47, aged 8–11 years) in two classes wore an accelerometer and sealed pedometer. Accelerometers were programmed to record at 15s epochs and the number of pedometer steps taken were recorded at the end of each school day. Patterns of activity by total accelerometer counts, and with cutoffs applied, were examined against time and the number of steps taken.Based on the accelerometer (>1500counts/min), the majority (68%) of children spent more than 60min in moderate to vigorous physical activity (MVPA) during the school day. Overall, time spent in MVPA was three times greater during break times compared to classroom time, with no apparent differences by class/age or gender. Total accelerometer counts and pedometer steps were correlated (r=0.60). One minute of MVPA was equivalent to 45 (95% CI 20 to 70) pedometer steps.The pedometer, a practical, relatively inexpensive tool, is suitable for the identification of less active children and has some utility to assess the relative intensity of activities in the context of a school day.