Exploring Nonlinear Effects of the Built Environment on Employment Behavior Among Older Adults: Evidence from Metro Station Catchment Areas

Social interaction, such as voluntary employment, can promote well-being and mental health for older people. Since walking and public transit are two major commuting modes for older adults, understanding the determinants of older employment behavior near metro stations is critical for the government and urban planners to encourage older employment. Using the mobile signaling data of 1,640,145 older employees and other multi-source spatiotemporal datasets in Shanghai, the Light Gradient Boosting Machine (LightGBM) is employed in this study to explore the nonlinear effects of the built environment on older employment near 333 metro stations. Results show that density, diversity, and design variables have a significant contribution on older employment, while distance to the city center, employment density among all age groups, and the number of older residents are the top three important variables. Partial dependence plots reveal that all independent variables have irregular nonlinear impacts on older employment. Each variable needs to reach an associated threshold to maximize older employment, and their nonlinear impacts are only effective when they are within certain ranges. Research findings can promote older employment and benefit mental health among older people by helping the government prioritize urban planning policies or interventions.