ABSTRACT
This conclusion presents some closing thoughts on the concepts covered in the preceding chapters of this book. The book discusses the various possibilities of categorizing traffic accidents to achieve optimal power. It examines the issue of stability of accident record over time, and a meta-analysis of available evidence showed that the previous conclusions regarding this problem have been mistaken. The book analyzes the matter of culpability for crash involvement, regarding how it is conceived and used by researchers and what the consequences are for the results. It investigates the commonly accepted but very peculiar 'fact' of a curvilinear association between exposure and traffic accidents. Many traffic accident prevention measures are based on opinions, beliefs, suppositions, hunches, and 'common sense', despite the fact that valid evidence often is available. It is difficult to envision how research into accident categories should be undertaken, apart from tests of differentially constructed accident variables as outcome measures.
