Social capital and perceived health of three types of older rural-to-urban migrants: A machine learning–based analysis

The transition from a familiar rural lifestyle to an unfamiliar urban setting can instigate substantial shifts in the social capital of migrants, potentially influencing their perceived health. Using data from the 2015 China Migrants Dynamic Survey, we analyze the relationship between three types of social capital—bonding, bridging, and linking—and perceived health through logistic regression and interpretable machine learning techniques. Our analysis focuses on older migrants with differing motivations: employment-seeking, grandchild caregiving, and city-based aging. Findings reveal that increased social capital generally enhances perceived health, with bonding and linking social capital particularly benefiting migrants with family and employment-oriented motives. However, this positive effect is not evident for grandparent caregivers with temporary urban stays. These results underscore the nuanced influence of migration motivations on social capital’s impact on health, offering valuable insights into the health dynamics of China’s aging migrant population.