The modifiable areal unit problem in the relationship between jobs–housing balance and commuting distance through big and traditional data

Abstract

Jobs–housing balance aims to reduce commuting distance and alleviate traffic congestion as a transport policy. However, the measurement of jobs–housing balance is affected by the spatial areal unit employed in analysis and planning. The mechanism by which commuting distance is affected by jobs–housing balance remains controversial for several reasons, including inconsistency of spatial analysis units and whether socio-demographic attributes are controlled. Cellphone data exhibit superiority in identifying the journey-to-work trips for large sample sizes and entail high spatial precision that can be integrated into different analysis units. By combining cellphone data and traditional survey data with detailed socio-demographic attributes aggregated at different analysis units, this study examines the scale and zoning issues of jobs–housing relationships in Shanghai, China. When using large analysis units, the effect of jobs–housing balance was expanded and that of different socio-demographic attributes changed variously and unpredictably. Furthermore, the uniformity of the spatial analysis unit influenced the effect of jobs–housing balance on commuting distance. This study substantiates the effect of the modifiable areal unit problem (MAUP) on the relationship between jobs–housing balance and commuting distance considering socio-demographic attributes and demonstrates that the changing rules of the effects of that balance on the commuting distance are different from those of socio-demographic attributes as the scale changes. Consequently, this work enriches the literature on the MAUP in relation to the jobs–housing balance and obtains a thorough understanding of the MAUP effects through combining big and traditional data.