This paper describes some of the results of an investigation to find the relation between the number of driving faults recorded during the driving test and accidents in the first six months of post-test driving. Subjects completed a questionnaire. There was no obvious pattern of accident involvement being associated with high numbers of driving faults. Women are less likely than men to be involved in accidents. People who made more faults tended to be older and more cautious drivers. Mathematical models were fitted to the data. Restricting analysis to non-low-speed manoeuvring accidents produced results similar to those for all accidents. Driving faults were categorised and statistical modelling used again. Reasonably strong predictive relationships between driving faults and accidents are revealed. The survey provides no good evidence to support the setting of separate failure criteria for subcategories of driving fault. For the covering abstract see ITRD E116025.