Head injury is the primary cause of pedestrian fatalities. To address this problem, the simplified sub-system test procedure has been used to distinguish vehicle stiffness in recent decades. Compared to the full-scale dummy test, the sub-system test has the advantage in terms of cost, repeatability, and testing coverage. However, the sub-system test methodology has long been criticized for oversimplification and for missing headneck interaction, vehicle pre-deformation, and head rotational kinematics, which make the ability of the component test to represent the real full-scale impact questionable. As a result, a potential problem may exist because vehicle design is optimized towards the sub-system test but may not necessarily provide the best protection in real pedestrian accidents. The goal of this dissertation is to study the relationship between the impactor responses of sub-system test with the head response in full-scale pedestrian impact. This process involved multiple steps:​ 1) evaluate the biofidelity of the PFEM;​ 2) compare the impactor response in the sub-system test and head response in the full-scale impact;​ 3) test the hypothesis that vehicle design variations have different influences on the impactor response and head response;​ and 4) investigate the factors that may contribute to the difference between the component test and the full-scale impact. The results of this work demonstrate limitations in the sub-system test that prevent it from reproducing the full-scale head response, even with matched input kinematics. Many of the limitations of sub-system tests were focused on the lack of vehicle hood pre-deformation caused by the pedestrian torso. The methodology and results of this dissertation could provide reference for improvements to future vehicle safety design and pedestrian regulation development.