Abstract
This research aimed to investigate the key factors that can affect the elderly's choice of access mode to the subway in the metropolitan area. It was conducted with two sequential parts. First, the respondents to the survey were categorized to three groups (the aging and aged commuters, aging and aged non-commuters, and general commuters) according to age and travel purpose, and their relationship was analyzed. Secondly, the hypotheses related with the elderly's choice of access mode to the subway were verified using a discrete choice model of a stated preference technique. An actual proof analysis was accomplished using the MDC procedure of the SAS statistics package and a deduced utility function, which sets the choice behaviors of the elderly's as the dependent variable and the surveyed factors as the independent variables. Through changing the conditions of the access mode choice the function predicts the variations that occur in the demand for the relevant mode and other modes. The demand can be determined through an analysis of its elasticity to the variations of each attribute. It is expected that the results are able to contribute to promoting the elderly's mobility even in an aged society which will occur to us in the near future.