Full‐scale rigid barrier crash tests were conducted at 40, 48 and 56 km/h to investigate the responses of seat belt restrained Hybrid III 5th percentile and Hybrid III 10‐year‐old anthropometric test devices (ATD) in motor vehicles. The 5th percentile responses were evaluated as a function of seating position in the vehicle and the seat track location. Hybrid III 10‐year‐old responses were evaluated in rear seating positions with and without booster seat installations.

Impact speed had little influence on the chest deflection of the front seat occupants but was found to be positively correlated for rear seat occupants. Head, neck and chest responses of the 5th percentile were consistently higher for rear seating positions. At least 45% of the 76 tests conducted with the 10‐year‐old with and without a booster seat had a HIC36≥1000 in the absence of head impact, while approximately 80% surpassed the neck injury assessment reference values. Chest deflection responses were elevated in tests where the torso belt was correctly positioned on the shoulder while chest acceleration appeared unrelated to the kinematic responses observed during the crash sequence. Implications for the future evaluation of rear occupant safety and booster seats are discussed.