This study investigates driver response attitudes to traffic information provided through variable message signs (VMS). It develops VMS driver response models using stated preference data collected through three different survey administration methods: an on-site survey, a mail-back survey and an Internet-based survey. In the process, it highlights the strengths and limitations of each method in eliciting driver response attitudes to information provision. The use of different media for the

more »

... ey administration provides insights for the design of travel surveys. A key study focus is to evaluate the effectiveness of Internet-based surveys for analysing driver behaviour under information provision. The results illustrate that a combination of survey administration methods may generate more representative data. They also indicate a high correlation between VMS message type and driver response. This suggests message content as a control variable for traffic system operators to trigger optimal routing policies under congested conditions to improve network performance. The benefits afforded by Internet-based surveys in the study context are highlighted. They are costeffective, amenable to automation, less labour-intensive, can target certain market segments more effectively and can enable greater clarity in the survey through better visual articulation. However, their widespread use requires greater market penetration in terms of Internet access.