Clavicle fractures are common injuries in three-point belt restrained occupants involved in frontal and lateral car collisions. Therefore, better understanding of clavicle loading which occurs during an impact and clavicle structural/material properties could help in the optimization of seatbelt restraint systems. Six clavicles from three post mortem human subjects were tested in a three point –bending test setup with pinned-pinned boundary conditions. The clavicle extremities were fixed into potting cups which were able to rotate freely about a single rotational axis (inferior-superior axis) and then, were loaded in anteriorposterior direction by an impactor at the middle shaft level. Two tests were performed on each clavicle:​ a) A non-injurious quasi-static test (1mm/s impactor rate) up to approximately 400 N b) A dynamic test (1m/s impactor rate) to failure. Reaction forces and moments were measured at both clavicle supports. A finite element model of a human clavicle was developed and used to simulate the tests. The results showed an averaged clavicle stiffness of 214±34 N/mm in the quasistatic tests. Concerning the dynamic tests to failure, the average maximum force was 1159±133 N, the average maximum deflection was 4.9±0.7 mm and the average maximum strain was 1±0.2%. The most common failure location was the middle third of the bone, which is consistent with literature data. The FE results showed a good agreement with test data. In addition to providing validation data for computational human models and dummies, the results of this study may lend insight into the development of advanced belt restraint systems.