Self-selected walking speed is being increasingly used as a primary outcome measure in the management of neuromuscular disease. It would be useful if the speed recorded in the gait laboratory represented the child's walking speed in the community. This study investigated the difference in self-selected walking speeds between a 10-meter walk, as measured during instrumented gait analysis, and a 10-minute walk. The authors found that self-selected walking speed during the 10-minute walk was slower than the self-selected walking speed recorded during the 10-meter walk. The former may be more representative of walking speed in the community setting. Walking speed measured during walks of 10 minutes or more should become an integral part of gait laboratory evaluation.