Consumer travel data were taken from the 1996 and 2001 versions of the Transportation Tomorrow Survey (TTS). The TTS is a cross-sectional weekday survey of the travel behaviour of approximately 5 percent of the households within the Greater Toronto Area and beyond (the list of participating municipalities has changed over time). Conducted every five years since 1986, the survey is typically carried out in the fall. Since 1991, travel data have been recorded for persons aged 11 years and older living in participating households. The data are then expanded using factors derived from Census occupied-dwelling counts, to estimate total population and travel behaviour across the survey area (JPT, 2003a, 2003b).
While the TTS provides a reliable sketch of weekday travel behaviour, it cannot be used to study travel behaviour on weekends, which is characterized by greater non-work travel (including shopping), longer trip distances, and lower transit shares (Kitamura and Van der Hoorn, 1987; Hu and Young, 1999; Bhat and Gossen, 2004; Bhat and Srinivason, 2005; McMillan et al., 2005). Moreover, it does not capture seasonal fluctuations in shopping frequency, duration, schedules, locations, and levels of cross-shopping3 (Roslow et al., 2000; Peter and Olson, 2004). There is no extensive longitudinal record of activity-travel behaviour available for the GTA. Furthermore, the TTS does not systematically record data for non-motorized modes of transportation (e.g., walk, bike, etc.), other than for trips to work or school.
Data from the TTS do suggest, however, that travel for weekday shopping is not only increasing, but growing more rapidly than travel for other purposes (Buliung et al., 2007). It is therefore important to understand the relationship between changes in the spatial structure of the retail economy and the design of retail destinations, and weekday consumer travel behaviour, in order to develop policies to accommodate consumer travel demand within the region's "sustainability" agenda. Moreover, knowing when and how consumers carry out weekday shopping is essential to understanding the competing needs of road users and some of the causes of congestion, and for finding ways to spread the demand for road use across the entire day.
A further limitation of the TTS is that the use of automobiles for non-work trips (such as shopping) in non-peak hours tend to be under-reported. Evidence from the United States, for example, points to an increasing share of this form of travel (Gordon et al., 1988; Handy et al., 2002; USDOT FHWA, 1995; USDOT BTS, 2003). Moreover, data on trips that start or end at home over a 24-hour period suggest that since 1986 in the GTA, discretionary trip-making has increased at a faster rate than that of trips to and from work or school (JPT, 2005). The under-reporting of these discretionary trips occurs in part because the TTS asks one adult per household to report the travel behaviour on behalf of all household members (JPT, 2003a). Estimates of consumer travel in this study therefore likely understate actual levels of weekday shopping in the GTA.
Finally, analyzing consumer travel behaviour over time using travel data for traffic zones requires a stable and consistent delineation of those zones during the period of interest. The TTS zone system, however, has changed from one survey period to the next, responding to changes in the transportation system, population growth, and the addition of new participating municipalities. To get around this difficulty, we have aggregated TTS micro-data (that is, data for areas smaller than zones) to the 2001 system of TTS traffic zones.
"Cross-shopping" means shopping across retail categories, usually at multiple stores, within a single trip.