Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
33 BREAKOUT SESSION Survey Methods Eric Petersen, RAND Europe Peter Vovsha, PB Consult, Inc. Stacey Bricka, NuStats Partners, LP Chandra Bhat, University of Texas at Austin Bruno Kochan, Transportation Research Institute, Hasselt University, Belgium Tom Bellemans, Transportation Research Institute, Hasselt University, Belgium Davy Janssens, Transportation Research Institute, Hasselt University, Belgium Geert Wets, Transportation Research Institute, Hasselt University, Belgium DIRECTIONS FOR COORDINATED IMPROVEMENT OF TRAVEL SURVEYS AND MODELS Eric Petersen and Peter Vovsha Peter Vovsha discussed data requirements to support the estimation process of activity- based models and improvements to travel surveys. He described the demands of the new models and promising areas of research related to travel surveys. Volume 2 includes a paper on the topic.1 The following points were covered in his presentation. ⢠Household travel surveys remain the major source of data needed for activity- based models. The basic sur- veys required for activity- based modeling applications are similar to those required to update and revalidate conventional models, although some additions are desir- able. The development of the new generation of activity- based models has provided the opportunity to examine the various types of travel surveys. This examination has identified some data inconsistency not previously noted. The development of new models has also created demands for new data and possible changes to travel sur- veys. Trips represent the unit of analysis in traditional four- step models. The units of analysis in activity- based models include trips, tours, activity episodes, and time allocation. ⢠Household surveys were conducted as part of the development of new activity- based models in New York, Columbus, Atlanta, and the San Francisco Bay area. The New York survey included approximately 11,000 house- holds in a 1-day survey. The Columbus survey was also a 1-day effort involving 5,555 households. Two- day sur- veys were conducted in both the San Francisco Bay area and Atlanta, covering 15,064 households and 8,069 households, respectively. A review of these surveys iden- tified concerns related to missing and miscoded loca- tions, in- home activities, conflicting joint activities and travel, underreporting of multiple activities, underre- porting of nonmandatory activities, and underreporting of preschool children. One example of underreporting relates to the percentage of workers making at- work sub- tours for lunch, banking, shopping, and business activities. ⢠Conducting on- the- spot checks represents one approach to improving household survey results. Items to check include consistent trip time locations and modes and consistent departure and arrival times. Joint travel by drivers and passengers can be checked for intrahouse- hold trip synchronization and interhousehold trips with colleagues, friends, relatives, and casual carpoolers. Joint intrahousehold and interhousehold synchronization can also be monitored. The presence of routine activities 1 See Petersen E., and P. Vovsha. Directions for Coordinated Improvement of Travel Surveys and Models. In Conference Proceedings 42: Innovations in Travel Demand Modeling, Volume 2: Papers, Transportation Research Board of the National Academies, Washington, D.C., 2008, pp. 85â88.