Geographic extent of income-based residential segregation between and within the school districts in Suzhou, China

Recent urban studies have highlighted the significant role of education in the underlying dynamics of urban segregation, particularly residential segregation. In China, residence-based education policy links school enrollment to place of residence, driving up housing prices in desirable school districts and inducing residential segregation. However, little research has been conducted on the extent of residential segregation, specifically in school districts. We used multiple data sources and the multilevel index of dissimilarity (MLID) to measure residential segregation within and between different school districts. Additionally, we employed a two-stage least squares (2SLS) model to address endogeneity issues related to school quality and residential segregation. Our findings revealed significant segregation, especially between high- and low-income families. Higher school rankings intensify segregation among high-income families, whereas school popularity decreases it among low-income families, potentially leading to gentrification. Although the score-based enrollment policy aims to integrate migrants, it also increases segregation among low-income families. In an involuted society, segregation is likely to increase if the disparities in school quality persist.