Assessing clinical change in motor function in children with cerebral palsy is a complex measurement task. Whereas a variety of methods have been developed to quantify specific aspects of gross motor behavior (eg, gait analysis, electrophysiological tests, energy-consumption techniques), systematic measurement of overall gross motor function is a more difficult problem. This special communication reviews the structural and performance characteristics required of a well-developed, valid, and responsive clinically based evaluative measure. We discuss several recent approaches used to assess responsiveness and critically examine clinical measures used in randomized controlled trials of physical therapy for children with cerebral palsy. It is argued that the creation and validation of responsive evaluative measures is essential if we are to assess accurately whether our treatments do more good than harm.