The aim of this paper is to report the results of an analysis of non-response, using what is believed to be a unique dataset collected in the context of the VITAL project undertaken by the Transport Research Centre at the University of Melbourne. This dataset consists of information collected in a validation survey in which intensive in-depth interviews were undertaken with both the respondents and the non-respondents who had taken part in a more conventional mailout/mailback survey. This dataset permits a systematic identification of the factors influencing both the magnitude and character of unit and item non-response. The first section in the paper discusses, within a general framework, the problem of framework, the problem of non-response and briefly reviews the existing evidence from the transport literature and elsewhere. The second section goes on to describe a modelling system which allows the joint investigation of unit and item non-response, including the investigation of survey administration factors such as the number of reminders sent to potential respondents. The third section briefly describes the VITAL dataset which consists of in-depth interviews conducted with almost 700 initial respondents. The fourth section presents the results of the analysis of non-response and discusses their significance in the context of current survey practice. The final section presents some overall conclusions from the work and on the basis of these conclusions, highlights a number of important directions for future methodological research. (A) An abstract only is published in the seminar proceedings.
Abstract