Using Existing Stated Preference Data to Analyse Bus Preferences

The purpose of this research is to re-analyse existing Stated Preference (SP) data sets which provide information on preferences towards bus in order to complement analysis commissioned by DETR which is concerned with bus fare elasticities and which is conducted at a much more aggregated level. The data sets available relate solely to urban/suburban travel and to studies for which ITS had prime responsibility. The objectives of this re-analysis of previous bus users' SP data sets were to address the following issues:

• How the sensitivity to cost varies with income, age and gender in addition to journey purpose. Analysis of the effect of income on the cost coefficient is essentially the same as analysis of how the value of time varies with income.

• Whether there is a non-linear relationship between the sensitivity to fare and the level of fare.

• Whether the sensitivity to fare varies according to whether the fare level in the SP exercise implied an increase or reduction on current levels. This would follow along the lines of similar analysis conducted on the sensitivity to time losses and gains in several recent value of time studies.

• How the magnitude of the fare change impacts upon the sensitivity to cost. This can be examined given that we are in a position to deduce whether the SP exercise implied an increase or reduction in the fare.

• Whether the quality of service offered influences the sensitivity to cost variations; for example, bus users may be less sensitive to cost increases where the service quality is good.

We had proposed to examine the reactions of both bus users and car users to changes in bus service quality and fare. However, it turned out that the car users’ data set, which provided information on car users’ preferences towards various bus attributes, could not be analysed because some variables in the data set could not be identified whilst details of only four of the twelve SP exercises used could be recovered.