Understanding the modifiable areal unit problem and identifying appropriate spatial unit in jobs–housing balance and employment self-containment using big data

Abstract

Jobs–housing balance (JHB) and employment self-containment (ESC) have been used to examine the jobs–housing relationship. However, the effect of the modifiable areal unit problem (MAUP) on ESC and JHB has received little attention. This study aims to examine the effect of the MAUP on the spatial variation of ESC and JHB by utilizing mobile positioning data from Shenzhen, China. Journey-to-work trips are examined at the individual level and then aggregated into different spatial areal units. It is found that the average ESC increases with the increase in spatial areal units and that the relationship between JHB and ESC is amplified when the spatial areal unit increases with spatial aggregation. A 2 km grid is found to be an ideal spatial unit for the analysis of ESC in Shenzhen because it is the turning point in which the increase in ESC started to slow down, and the decrease in the coefficient of variation began to diminish. In addition, workers were more likely to commute by non-motorized transport modes when their jobs were within 2 km. This study helps elucidate the effect of the MAUP on ESC and JHB as well as determine the appropriate grid size for analysis. This study further suggests that the ideal spatial unit for the analysis of ESC and JHB may be related to the transport mode of the city under study.