This study examines the ways in which urban context affects vehicle trip-generation rates across a variety of land uses. An establishmentintercept travel survey was administered at 78 establishments in the Portland, OR, region during the summer of 2011. Data were collected from high-turnover (sit-down) restaurants (Mexican and pizza), 24-hour convenience markets, and drinking establishments. Combined with persontrip counts, vehicle-trip counts and built- environment data, a method to adjust Institute of Transportation Engineers (ITE) vehicle-trip rates to reflect a local community’s context has been developed. Results from this study reveal a trend: For all land uses tested here, vehicle-trip rates decrease as neighborhood types become more urban. Comparisons between ITE trip-generation rates and vehicle-trip rates from this study indicate a need for a local adjustment for both convenience markets (open 24-hours) and drinking establishments. High-turnover (sit-down) restaurants are consistently predicted by the ITE methodology, but based on our findings we recommend a vehicle-trip rate adjustment to better match locally observed travel patterns. A model to adjust ITE’s trip-generation rate for urban contexts was developed in this study. The key measure representing urban context is the average Urban Living Infrastructure (ULI) score from the Metro Context Tool within a half-mile buffer around establishments. ULI is a measure representing the density of retail and service establishments serving daily needs, and is highly correlated with other built-environment measures such as lot coverage, density and accessibility to transit. The model developed here has a good statistical fit and ease of use in an evaluation of new development. The approach is also useful in guiding plans as we have related the ULI measure to other planning-relevant, built-environment measures. The study findings are limited in a number of ways. The three land uses examined and the relatively small sample size limit the number of factors that could be accounted for in our statistical analysis. In addition, data collection was limited to the weekday, evening peak hour of the facility for each of the three land uses. The findings are localized and may not have broad applicability beyond the Portland region. Work planned for the immediate future includes validation of the method using data collected from additional sites in Portland and elsewhere, and analysis of site-level attributes that include parking, building orientation, pedestrian and bicycle infrastructure, and other design features.