An exploratory data analysis of fatal and serious motorcycle collisions within Devon between 1996 and 2001

Published by Devon County Council

August 2003

ISBN 1-85522-894-7

Full Report (1.01MB - pdf help)

Executive Summary

This work is intended to make an initial exploratory data analysis of STATs 19 information recorded on motorcycle collisions within Devon where there was either a fatal or seriously injured motorcyclist. Where possible, this is contrasted with results reported in the extensive research literature on motorcycle collisions and causations. It is hoped that this will provide a framework at least for identifying where further research effort should be directed.

Some tentative suggestions for possible countermeasures, based more on the literature than the local data, are also made.

Some of the most striking observations, either from the literature or from the collision data are as follows:

• Informal training with the sole aim of increasing experience with a given machine may have a value.
• Targeted training at teenagers / moped riders is clearly necessary. Whilst it is unclear how this may be made attractive to those selecting scooters on the basis of fashion, those seeking cheap mobility may be more amenable to suitably priced training of a appropriate type. Some people in this age group are provided scooters by statutory bodies, it may be possible to oblige some additional training as a condition of the provision of the machine.
• There is strong evidence that the peak age for fatal casualties is somewhat high (30 - 40 year olds, and higher than the corresponding peak for serious casualties, and there is evidence of leisure riders being over-represented. This is being researched by Leeds University so discussion with them may be beneficial before taking this work too much further. One obvious piece of supplementary information would be to go through the coroner’s reports to determine the type of bike involved in these collisions. This is something that has changed over time, for example Woodward (1983a) demonstrated how motorcycle fatalities used to peak around 20 years of age.
• Careful consideration needs to be given to the Integrated Transport possibilities of Two Wheeled Motor Vehicles (TWMVs). It appears that riders could do a lot more to improve their own safety, which implies that careful riders do not experience quite the same level of risk as the RAGB overall estimate of risk for motorcyclists. It should be noted that the officially estimated risk per kilometre for motorcycles is currently lower than for pedal cycles, and we are working very hard to promote pedal cycling. Wilson (1984) also argues that the overall car collision rate includes a higher proportion of lower risk drivers than the motorcycle collision rates, which are dominated by high risk young riders. The risk to a trained and cautious motorcyclist could therefore be much lower than the overall figure suggests.
• There is scope for further work on classifying roads where fatalities take place. For example, the IT available does not yet allow geographic profiling of single vehicle loss of control collisions yet it may be possible to highlight areas of the network that perhaps should be avoided by inexperienced riders or which require a specific road safety intervention.
• There is still scope for dealing with “looked but did not see” collisions, and any efforts in this regard would affect more than just TWMVs. There are claims that police drivers can be taught to look and see a little better, this therefore raises the question of whether a flavour of this training can be introduced into any course offered in any context by Devon County Council, and also whether more general educational activity can be directed to this end.

It is hoped that this work will stimulate discussion with partners and other stakeholders who have experience of motorcycling, training, collisions or behaviour. This data analysis can be modified in the light of further questions that may arise, in particular it may benefit from contrasting matched motorcyclist demographic groups with car-drivers, and it may also prove necessary to include a random sample of slight collisions. One thing that is very clear from the data analysis is that there is not a single homogenous group of “problem” motorcyclists. There appears to be a wide range of styles, behaviours, attitudes and machines. It could be extremely unhelpful to extrapolate opinions about how to deal with a perceived problem group (such as the "born-again biker”) with known problem groups (currently young riders on mopeds).