Analysis of Latent Classes by Survey Response Durations in a Driving Behaviour Simulator with In-Veh

• 作 者: GuoJingqiu;QiuMin;
• 机 构: The University of Western Australia M 261,35 Stirling Highway,Crawley WA 6009,Australia
• 出 处: 第八届全国交通运输领域青年学术会议
• 发表日期: 2009年期

Intelligent Transportation Systems represent the application of advanced technologies to transport in order to improve system efficiency and safety.A computer-based stated-preference (SP) simulator was developed to investigate how drivers would respond to innovative in-vehicle information when they are driving with an ecodriving support device.The sample collected online through the SP simulator has a size of 283 responses and the sample is unbiased as far as age and gender are concerned.The survey response duration (SRD) is also recorded as secondary information but is important for understanding the number of latent classes that is a way to deal with the poptflation heterogeneity within the family of logit choice analysis.It is found that the SRD follows unexpectedly a log-logistic distribution and two or three latent classes are suggested in the estimation of a set of latent class models for the study.