Document Type


Publication Date



Traffic surveys -- Oregon -- Portland, Trip generation -- Oregon -- Portland, Transportation -- Planning -- Statistical methods, Urban transportation -- Environmental aspects


In the past decade, the methods for estimating multimodal transportation impacts of urban land use development have improved substantially. One assumption commonly made in these new methods is that overall person-trip rates at similarly-sized establishments of the same land use do not vary across a region. This is an assumption of convenience to permit the adjustment of ITE Trip Generation vehicle trip rates for use in different urban environments. However, this assumption is inconsistent with theories of urban economics, which recognize that businesses pay a premium to locate in areas with high levels of accessibility to attract more customers. In addition, most transportation impact analyses have ignored income effects, even though socio-economics are a proven driver of travel behavior. To test this assumption and understand the effects of accessibility and income on levels of activity at the establishment level, we examine transaction counts for 97 grocery and convenience markets in Portland, Oregon. In a multilevel negative binomial regression, we test the relationship of regional accessibility, local accessibility, and income on weekly and daily transaction rates. While there was not enough evidence to suggest a significant relationship between accessibility and transaction rates, the results indicated a significant relationship with median income of the surrounding area. The implications point to the need to consider area-wide socio-demographics in site-level transportation impact analysis. The study also provides some important discussion about the use of transaction as a proxy for person-trip rates.


Paper No. 17-03243

Poster was presented at the Transportation Research Board 96th Annual Meeting in Washington DC.

Persistent Identifier