The role of street network metrics in shaping distance distributions in a residential neighbourhood

Abstract

Walkability studies often rely on physical proximity metrics for destination accessibility. However, some recent studies argue that configuration (topological) distance provides a more accurate reflection of the actual distances experienced by individuals. This study therefore aims to test the extent to which an association exists between network-metric proximity to several destinations and two main measures of space syntax – integration and choice – in a dendriform network structure such as Putrajaya city (Malaysia). Using GIS, multiple buffer-service areas were generated around each housing unit (N = 2,392). Interpolation and space syntax analyses were conducted to assess metric and topological distance at local and global levels for each buffer. The statistical analysis showed that distances to commercial areas, transportation and average distances to all destinations were strongly affected by the syntactical properties of the neighbourhood. Moreover, global measures were more powerful in detecting metric-distance changes compared to local measures. Aligned with natural movement theory, these results support the idea that the spatial properties of urban form have a significant impact on distance distributions within residential settings, specifically distances to commercial areas. Therefore, policies that promote mixed-use development, especially in areas with good transportation access, should consider the impact of syntactical accessibility of the network.