A pilot investigation of driver error was carried out on roads near the laboratory to determine a) which were the most common errors, b) which were the most dangerous errors and c) the locations at which they occurred. One hundred and eight subjects drove around a 28 mile route accompanied by a trrl observer who monitored aspects of their driving behaviour. Subsequently, the behaviour of 354 drivers was observed from a moving car along sections of the same route. Over the same period, time-lapse photography was used at specific points on the route to observe a larger number of drivers, 2,800 of whom were recorded committing driving errors. Each error was allocated a score on a 5-point scale according to a subjective estimate of its level of danger. Eighty different types of driving error were recorded and classified in terms of frequency, severity and location. A high level of correlation was between (i) the observed errors together with their level of danger and the errors which led to injury accidents, and (ii) the locations of observed errors, and the locations at which injury accidents had occurred. (Author/publisher).
Abstract