Effect of land use on shopping trips in station areas: Examining sensitivity to scale

Abstract

Abstract Land use planning has been emphasised by both planners and researchers in providing incentives for local shopping. This study advances the body of knowledge by investigating land use and shopping trips in the catchment area for the metro station nearest to home, a unique context that serves both as a shopping destination and a transit origin for further shopping opportunities. Negative binomial and Tobit regressions were applied respectively to examine the effect of land use density on weekly shopping frequency in the station area nearest to home, and its share of individual’s total shopping. The impact of the Modifiable Areal Unit Problem (MAUP) on these statistical relationships was addressed, with land use densities objectively measured in seven spatial scales using a radial buffer scheme. Results from all regressions suggest the existence of MAUP in influencing the sign, significance and size of the station area land use effects on shopping trips. A nuanced disaggregation of service types identifies both complementary and competitive relations among services, even under the same generic category, such as dining. The results show that model adopting 400 m radial buffer as measurement scale accumulates greater effect of service land use on station area shopping frequency, while model adopting the 1200 m scale accumulates greater effect of service land use on the station area shopping’s share of individual’s total shopping. This paper presents the sensitivity of effect of land use on shopping trips in station areas to the spatial scale of land use measurement. It provides pragmatic guidelines for station area planning to augment retail attractiveness and the social benefits of the catchment area for the metro station.