This article considers methodological issues relevant to the study of differential crash involvement and reviews the findings of research in this area. Aspects of both driving skill and driving style appear to contribute to crash risk. Of the former, hazard-perception latency appears to play an important role, and this may be attributable to generalized abilities to identify visual targets in a complex background and to switch attention rapidly. Of the latter, faster driving speed and willingness to commit driving violations increase crash risk, and these factors may be explicable in terms of personality and antisocial motivation. The article concludes with an examination of the practical implications and of the ways in which research in this area might usefully proceed.