Objectives Spatial inequality refers to the unequal distribution of resources in space due to social and economic development. It can be manifested in various aspects such as the living environment and social services. At present, most of the research on spatial inequality is based on the qualitative description of static data. The data comes from traditional methods such as questionnaire surveys. This type of data has the disadvantages of high collection cost, long cycle, and slow update. In comparison, traffic data has become an important measure of spatial inequality on account of its accuracy and objectivity.
Methods Based on the census and traffic data of taxis in Shanghai, we conduct a spatial multi-scale division of the Shanghai area, utilizes Hadoop to remove dirty data and extracts origin destination points. After using social network analysis, we obtains travel convenience and travel distance based on floating car data in each area at various scales, which are used as indicators to represent spatial inequality. Spatial regression analysis helps to explore the relevance of spatial inequality according to demographic information.
Results There exists spatial inequality of taxi ride in Shanghai based on the following two reasons: (1) In terms of travel convenience, the areas with the highest centrality are concentrated in the main urban area and outer sub-central area; (2) As for travel distance, the average travel distance in the central urban area is relatively small, while in the outer area farther from the central city, the average travel distance is larger.The above-mentioned spatial inequality variables are related to the population structure. In terms of the relevance of travel convenience and population structure, areas with lower proportions of children, teenagers, youth, agricultural, non-local populations, and higher middle-aged, non-agricultural populations have higher travel convenience. In terms of the correlation between travel distance and population structure, areas with lower proportion of young, old, agricultural populations, and higher proportion of young and middle-aged populations have longer travel distances.
Conclusions Regions with a higher proportion of vulnerable groups (children, elderly, migrants, agricultural population, etc.) have lower travel convenience and longer travel distances, this correlation changes with changes in spatial location and spatial scale.