研究

我目前的研究沿两条主线展开。其一，应用计算社会科学、因果推断方法，探讨城市社区如何影响居民的健康与福祉；其二，考察宏观社会经济与住房、教育政策背景下城市社区的变迁。目前代表性成果如下：

* 通讯作者

计算社会科学、因果推断方法下的邻里效应研究

Mixing or Socioeconomic Status? A Quasi-Experimental Study of Neighborhood Composition and Subjective Well-Being
Zhang, Y., Huang, L., & Ye, Y. · Social Science & Medicine, 2026, 404, 119518

摘要

Mixed-income development, an urban policy strategy aimed at protecting disadvantaged residents from concentrated poverty, has yielded mixed findings on subjective well-being, challenging its policy rationale. This study revisits this association by drawing two analytical distinctions that have received less attention in prior research: first, between neighborhood mixing indices and neighborhood socioeconomic status (SES) as distinct dimensions of neighborhood composition; and second, between linear and nonlinear specifications of their associations with well-being. We exploit Hong Kong’s random allocation of public housing as a quasi-experimental design that attenuates residential selection bias. Using multilevel growth models on Hong Kong Panel Study of Social Dynamics data and focusing on public housing residents, we find that neighborhood SES is positively associated with subjective well-being, whereas neither income mixing nor housing type mixing indices show significant associations. Quintile-categorical specifications further reveal that the SES association is broadly monotonic, while the income mixing index follows an inverted U-shaped pattern. By disentangling neighborhood SES from mixing indices and examining nonlinearity within a quasi-experimental framework, this study clarifies which dimensions of neighborhood composition matter for disadvantaged residents’ well-being, with implications for the design of mixed-income housing policies.

Neighbourhood Cohesion and Subjective Well-Being Among Disadvantaged Residents: Causal Evidence From the Public Housing Natural Experiment in Hong Kong
Gu, R., Ye, Y., & Zhang, Y.* · Population, Space and Place, 2026, 32(2), e70222

摘要

Neighbourhood cohesion is suggested to improve residents’ subjective well-being, especially for disadvantaged individuals who are more likely to live in resource-deprived neighbourhoods. However, existing studies face a challenge in establishing a causal relationship between neighbourhood cohesion and subjective well-being due to endogeneity issues from residential selection. To address this methodological challenge, we use the random allocation of public housing in Hong Kong as a natural experiment. We analyse data from the Hong Kong Panel Study of Social Dynamics (HKPSSD) and focus on public housing residents. We find that individuals’ pre-determined socioeconomic and demographic characteristics are uncorrelated with the social cohesion of the neighbourhoods where residents are placed, which confirms the validity of our research design. Based on the design, we find that neighbourhood cohesion is positively associated with residents’ life satisfaction and happiness, while it is also negatively associated with mental distress. Moreover, neighbourhood satisfaction is a significant mediator between neighbourhood cohesion and subjective well-being. Our findings contribute to establishing a causal relationship in studies on neighbourhood effects and provide policy implications for improving well-being among disadvantaged residents.

From Disorder to Distress: Investigating Neighborhood Change and Mental Health with Streetview and Longitudinal Survey Data
Zhang, Y., & Miao, J. · Social Science & Medicine, 2025, 384, 118589

摘要

Neighborhood physical disorder, a prominent negative aspect of the residential environment, is expected to adversely affect mental health. However, the limited use of longitudinal designs with objective measures of physical disorder hinders the establishment of causal effects and the generalizability of findings, particularly in rapidly changing urban contexts. This study integrates Streetview data with the Shanghai Urban Neighborhood Survey from 2017 to 2019 to examine changes in physical disorder and their association with psychological distress in Shanghai. Analysis of Streetview data reveals a decline in physical disorder over the study period. Employing fixed-effects models and generalized propensity score matching, we find that this decline is associated with reduced psychological distress. The association is significant among residents with lower socioeconomic status (SES), while it is less pronounced among high-SES residents. These findings contribute to the understanding of the effect of physical disorder by providing more robust evidence from a rapidly transforming urban context. They also shed light on the effectiveness of residential environment improvements in Shanghai and other cities experiencing rapid neighborhood change.

Gentrification and Mental Health in Hong Kong
Zhang, Y., & Miao, J. · Chinese Sociological Review, 2025, 57(5), 545–557

摘要

This study investigates the association between gentrification and mental health in Hong Kong, where gentrification is mainly driven by government intervention and is widely supported by the public. We utilize a harmonized dataset that integrates the 2006 and 2016 Hong Kong population census data with data from Wave 4 (2017/2018) of the Hong Kong Panel Study of Social Dynamics (HKPSSD). Findings of the multilevel logistic models show that residents in gentrifying neighborhoods report a lower risk of mental distress compared to residents in non-gentrifying neighborhoods with similar socioeconomic characteristics. Furthermore, residents with higher household incomes tend to benefit more from gentrification. Results from causal mediation analysis suggest that neighborhood cohesion mediates the effect of gentrification on mental health. This study advances the literature on gentrification effects by offering empirical evidence from the relatively understudied East Asian context and examining the socioeconomic variations in these effects. We discuss potential policy implications for future interventions in neighborhood change processes.

宏观社会经济与政策背景下城市社区的变迁

Public Housing and Neighborhood Composition in Hong Kong, 1981–2016
Zhang, Y., & Wu, X. · Social Science Research, 2026, 133, 103276

摘要

Public housing is recognized for changing the neighborhood socioeconomic landscape. In this study, we investigate the impact of public housing projects on the neighborhood socioeconomic composition in Hong Kong, an affluent city characterized by economic inequality and social divide. By employing a two-way fixed effects model on data that combines neighborhood census and administrative records, we identify that the share and count of low-income households increased with additional public housing projects in the neighborhoods. Conversely, we observe crowding out effects on high-income households due to public housing projects. Such associations are stronger in neighborhoods with higher initial economic status. Moreover, the effect on neighborhood composition is only evident for new public housing projects, whereas renewed public housing projects had little impact. The study contributes to a more comprehensive understanding of the effects of public housing on neighborhoods and offers insights into designing public housing policies in a comparative perspective.

Canceling the Admission Priority of Private Schools Enlarges Housing Price Gap in Public School Districts: Evidence from Shanghai’s New Admission Policy
Zhu, P., Zhang, Y., & Wang, J. · Real Estate Economics, 2023, 51(1), 49–67 （共同第一作者）

摘要

In 2018, Shanghai implemented an admission policy that canceled the admission priority of private schools to promote education equity. Since this policy is a unique measure for adjusting private and public school competition and discouraging private school choice priority, little is known about the policy effects. In this research, to examine the impact of the new admission policy on the capitalization of public education quality, we apply boundary fixed effect and Difference in Differences (DID) analysis to housing transaction records before and after the policy. The admission policy on average led to an additional 2% housing price premium for every standard deviation increase in public school quality. However, this average increase in premium was mainly driven by elite (top 5%) school districts, where an additional 8.6% housing price premium was generated by the policy. Housing prices in nonelite school districts, on the other hand, demonstrated no significant changes. These results indicate that the policy enlarges the housing price gap among school districts with different education quality. Thus, rather than promoting education equity, this policy may overall worsen the housing affordability in good public-school districts and make access to quality education more exclusive.