A REVIEW OF BRITISH EVIDENCE ON THE VALUATIONS OF TIME AND SERVICE QUALITY

This study builds upon the review of in-vehicle time (IVT) values reported in Wardman (1997) and its principal objective is to conduct a review of British empirical evidence regarding the valuations of the following variables: \n \nwalking time \naccess time \nsearch time \nlate time \ndeparture time adjustments \ncongested travel time \npublic transport headway \npublic transport interchange \n \nA number of other service quality variables have not been included in this review for various reasons. In some cases the range of effects discerned by a variable varies dramatically across studies, with mode specific constants providing a good example, whilst in others, such as with travel time variability or railway rolling stock, the precise nature of the variable being valued varies widely. For other variables, there is the problem that there is as yet insufficient empirical evidence or else there are problems of commercial confidentiality. Examples include the valuations of terminal facilities, advance purchase requirements and safety and cleanliness factors, but hopefully these problems will diminish over time and large scale quantitative reviews can be conducted in the future. \n \nThe valuations are obtained from disaggregate behavioural models developed since 1980 which have been estimated to Revealed Preference (RP) or Stated Preference (SP) data. The proposal specified a maximum level of disaggregation according to journey purpose and mode. This is because other segmentations, such as by income or household type, are not conducted on a consistent basis and analysis of their effects would have involved considerable additional resources. Nonetheless, future review studies might tackle such issues. \n \nThe aim of the research is to review how the valuations of the time and service quality attributes listed above and to examine how they vary according to key socio-economic and trip characteristics. Two approaches are adopted to examine the influence of these variables.