Influences on road safety development: a study of the applicability of models

How road safety develops in the course of time is determined by changes in crash rates and exposure (mobility). SWOV carried out a study of developments in crash rates as well as exposure. This was done by using a time series analysis technique that is known as state space modelling. The most important results reported here are of a methodological nature. How can state space modelling help us to find explanations for road safety developments as a time series? In spite of an extensive systematic search for possible explanations, only a few were found. What we did find concerned the influence of weather, the Start-up Programme Sustainable Safety, and the introduction of the Free Public Transport Pass for students. An important learning result of using state space modelling was that when developing an explanatory model one should take a number of complicating circumstances into account: the limited availability of data (especially explanatory variables), the relatively short time series of the data, and the fact that many developments occur simultaneously. This report may be accessed by Internet users at: http://www.swov.nl/rapport/R-2006-29.pdf