In San Francisco, auto-ownership levels have been declining, new transportation options such as bike-share and ride-share have become available, the bicycle network has expanded, and transit service is constantly evolving. Travel behavior, too, is changing, but it is not clear whether these changes are in response to changes in the environment, or due to changes in demographics over time, or some kind of fundamentally shift in attitudes toward different modes of travel.
When we forecast future travel demand, we assume that each individual’s travel preferences remain constant from today to the forecast year. We understand that factors such as age, income, and other socio-demographic attributes affect travel preferences and mode biases. We use travel surveys to understand that relationship and build models to estimate travel choices from those factors. We then boldly assert that the relationship between socio-economic demographics is the same 25 years from now as it is today. In reality, we know these relationships change over time. In fact, we re-estimate our models periodically to account for it (and a number of other things, like shifts in population make-up, expansion of transit service, or new travel options like bike-share). A better understanding of these trends will help enlighten our discussion of future travel demand projections, and may help us develop an appropriate way to address changing preferences in the modeling process.
In the Bay Area, the California Household Travel Survey (CHTS) 2010-2012 provides an update to its predecessor, Bay Area Travel Survey (BATS) collected in 2000, 1996, and 1990. This longitudinal data allows us to look at travel preferences and behaviors over a 20-year period We will analyze these travel behaviors and preferences among demographic cohorts as those cohorts age. How do Bay Area residents in their 20s in 1990 compare to their more contemporary counterparts in their 40s in 2010? How do people in their 20s, 30s, 40s, 50s in 1990 compare to those in the same age categories now? We will identify and describe trends that emerge and discuss implications for modeling practices.