The paper examines the application of six vehicle safety rating systems to a common crash database, for the purpose of making a comparison of the rating results produced by each system and to develop an understanding of the differences which emerge. The rating results are compared based on rank order of crashworthiness of vehicle models, and relationships between each pair of results. Finally, the results with their respective confidence limits are used to classify each vehicle model as having inferior, not defined or superior crashworthiness, and the classification is used to compare the relative discrimination of the methods.