29

This paper proposes a microeconometric study of road accidents. We try to estimate a multinomial logit model to analyze the levels of severity of road accidents in Tunisia. Referring to the accidentology literature, the severity can depend on three components: the driver, the vehicle and its condition of use, and the infrastructure. These various components constitute the road traffic system and determine road safety. They interact at a given time and place to explain the occurrence and severity of an accident. Several quantitative and qualitative variables can be identified and measured to describe these components. The objective of this work is to show how and to which extent these variables affect the gravity of a road accident. We seek to estimate the probability that a driver with certain characteristics and leading in a given traffic environment, incurs an accident of a given severity level. The structure of the estimate is based on disaggregated data collected following the study of index cards investigation proposed by the National Observatory of Circulation (Tunisia). The released results show that all the variables are statistically and theoretically significant and explain severity of an accident to different degrees.Classification JEL : D01, C13, C25, C35, R41.