This paper describes a survey technique which has been used for the measurement of public transport ticket usage rates. The objective of the survey was to provide information to assist in the allocation of fare-box revenue to operators within a privatised public transport framework. The method has been shown to work successfully and to provide credible estimates of ticket usage. A principle methodological aspect of the survey was the need to recognise, in advance, that the sampling method was not truly representative, and that extensive weighting would need to take place in order to obtain unbiased population estimates. In order to do this, specific information needed to be recorded before and during the survey, rather than waiting for the analysis phase before realising the need for weighting data. Similarly, the use of multiple survey methods (ticket usage questionnaire and observational survey) was necessary in order to obtain an estimate of the effects of non-response bias. (A)
Abstract