Biofidelic anthropomorphic test devices (ATDs) improve injury prevention by evaluating the effectiveness of motor vehicle safety systems. Due to a lack of pediatric validation data, pediatric ATDS have been size-scaled from adult data; but studies show that child subjects experience different motion relative to adults that are not simply scaled down by size. Previous studies note age-based differences in relative maximum excursions, but no study has quantified the effects of age on trajectory shape. In this study, low speed (<4 g) frontal sled tests were conducted on 30 human volunteers from ages 6 to 30 years. Subjects were restrained by a lap and shoulder belt. Photo-reflective markers placed on anatomical landmarks of interest, including the head top, were quantified by a 3-D near infrared tracking system. Subjects received six repetitive trials. Head top landmark trajectories in the sagittal plane were modeled using a 4th order Bézier curve equation using an iterative least-squares algorithm. A principal component analysis on the control points indicated that the first four components accounted for 95% of the variance. A linear regression analysis demonstrated that the first principal component was significantly related to subject erect sitting height (p<0.001) and other correlated measures of body size. Using the resulting regression model, shorter sitting height was associated with greater head excursion and a more rounded trajectory, indicating greater neck flexion. To our knowledge, this represents the first application of these functional shape analysis methods to impact biomechanics data. The method provides a concise, effective quantification of trajectories that will be useful for specifying and evaluating ATD performance.