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.
89 Using Global Positioning System Data to Inform Travel Survey Methods Stacey Bricka, NuStats Partners, LP Chandra R. Bhat, University of Texas at Austin While the transportation community continues to work toward the long- term goal of using Global Positioning System (GPS) technology to produce higher- quality trip files, the reality is that the current method of random samples, telephone surveys, and travel logs continues to be used. Thus, for any given regional travel survey, trip underreporting will occur at some level. The research question that forms the focus of this paper is whether an analysis of GPS data collected as part of a regional travel survey can be used to minimize trip underreporting through improved survey methods. The focus is on demographic characteristics, travel behavior characteris- tics, and indicators of adherence to survey protocol that potentially impact trip underreporting. The results sug- gest that, while more research into this subject is war- ranted, there are specific, low- cost changes to the survey materials as well as to the interviewing process that can be made immediately to reduce trip underreporting. Ten years ago, the transportation community beganin earnest an investigation into the application ofGlobal Positioning System (GPS) technology to the collection efforts for travel survey data. The immediate focus of this technology application has been to improve the quality of travel survey data, with a long- term goal of eventually replacing respondent- reported data with travel details collected passively through GPS devices. The main application of GPS in regional travel surveys to date has been for auditing trip reporting, to determine the level of trip underreporting by vehicle drivers, and to develop appropriate correction factors for the data. Specifically, GPS has been used in 12 regional travel surveys: Lexington, Kentucky (1996); Austin, Texas (1997); California (2001); Los Angeles, California (2001); Pittsburgh, Pennsylvania (2001); St. Louis, Missouri (2002); Ohio (2002); Laredo, Texas (2002); TylerâLongview, Texas (2003); Kansas City, Missouri (2004); Reno, Nevada (2005); and in a pilot test for the upcoming Oregon statewide travel survey (2005). In addition, other GPS studies not directly linked to regional travel surveys have employed GPS for speed studies and in testing the development of trip tables solely from GPS data. For purposes of this paper, references to GPS studies refer to those conducted as part of regional travel surveys only. In the conduct of these studies, several important facts have been gleaned: ⢠Respondents who self- select to participate in GPS travel studies are different from those who do not elect to participate. As documented in several travel survey reports, GPS participants tend to report higher incomes and own their own homes compared with those who elect not to participate (see, for example, NuStats). Thus, most of the findings to date and conclusions about trip underreporting are based on a select group of respon- dents and not general populations of entire regions. ⢠The methods used to process the GPS data streams vary across the GPS studies conducted to date and influ- ence the degree of trip reporting detected. Some studies, such as the Los Angeles study, used in- vehicle devices to capture trip details for both drivers and passengers, while others focused only on drivers. In addition, as shown in an early analysis of the Austin data, the time thresholds used in vehicle movement detection can cause the trip
