Abstract
If only 60 or 70% of all traffic victims are tested for alcohol, the results show a higher percentage with high blood alcohol levels than would actually be found if all victims could be tested. This is because the victims not tested include children, those involved in accidents in which alcohol is not believed to be a factor, and those who survive a sufficient period of time after the crash so that a blood test would no longer be considered useful. Other discrepancies between official and popular belief and results of research on alcohol involvement in traffic accident are also presented.