The frame of reference for analysing road user behaviour comprises the systems traffic environment and traffic participant, interconnected by the system traffic behaviour. Swedish studies have found a negative correlation between road resurfacing and accident rate. The reasons why accidents increase are: greater feeling of safety and comfort, greater ease of driving etc. Safety engineers concentrate on events immediately before an accident; it is contended that certain underlying factors such as the following have much greater significance: compensation of risk (drivers behave more carelessly as their feeling of safety increases); delegation of responsibility if certain problems are solved by the `system' such as fog and congestion warnings etc which produce tendency to rely on the system; speed/behaviour transfer - high motorway speed tends to continue off the motorway; behaviour diffusion/imitation - conformance to surrounding driving patterns; interaction problems - isolation from other road users. Interviews after accidents tend to provide biassed information. Conflict studies combined with interviews seem to have advantages: the cost is lower, conflicts can be collected with great reliability, interviews are conducted immediately and interviewees are not in shock as after an accident. The technique has high validity; it produces as valid predictions of expected accidents as past accident records.
Abstract