Road accident risk: an investigation into various assessment methodologies

It has been argued that the major cause of road accidents is due to
human error, and that drivers modify their behaviour on the road
according to the level of risk they perceive in the road environment.
Unfortunately, due to a lack of suitable methodologies no reliable
method for assessing drivers t perception of hazards has yet been
developed. For this reason several techniques for assessing perception
were investigated.
Non-directive, focussed and critical incident interviews, Q-sort and
several variants of the repertory grid were used in an attempt to
discriminate between old and young male drivers' perception of road
hazards.
Only the repertory grid discriminated successfully and in all variants
old drivers more often used extremes of the rating scale whereas young
drivers more often used mid-scale ratings. It was hypothesised that the
extreme responses of the older drivers' reflected their decisiveness,
which may arise from their greater experience and confidence. The
tendency towards mid-scale ratings for the younger drivers was
attributed to their lack of certainty in judgement.
The most successful repertory grid variant, namely the fixed repertory
grid, was used on four different groups of young and old drivers and
this work revealed further differences between the age groups in
correlational structures, element clusterings, and principal component
variances. The scales (and the hazards) that were significantly
correlated and clustered together, and the structure of the components
produced, showed consistency over all four groups.
Further work to develop the repertory grid is suggested and potential
applications are discussed. These include using the fixed repertory
grid, as a diagnostic instrument for detecting deficiencies in road
layout which lead to accidents at blackspots, and as an aid to driver
training programmes.