Abstract: Employing data from CLHLS survey conducted between 2005 and 2018, this study analyzes Chinese older adults' multidimensional poverty status and changes using the AF method. This work also summarizes the multidimensional poverty pattern by utilizing the latent class analysis and estimates the impact of family and social support on different patterns of poverty. The results suggest that the older adults' multidimensional poverty index declines with an increasing contribution of health. Multidimensional poverty patterns include financialmental poverty, comprehensive health poverty and mobilityparticipation poverty. Older adults who are female, older, and have lower education show higher risks of falling into three kinds of poverty. Older adults who are rural, live alone, and lack family and social support are more likely to be in financialmental poverty. Comprehensive health poverty is more likely to be observed among older adults living with family and have relatively sufficient community health services and social pension security. Older adults with less family support, community services and social security are more likely to show mobilityparticipation poverty.

摘要： 基于2005~2018年中国老年人健康长寿影响因素调查数据，采用AF测量方法，分析老年人多维贫困状况及其变化趋势；运用潜类别分析法，探讨老年人的多维贫困模式，评估家庭和社会支持因素对不同贫困模式的影响。研究发现，2005~2018年间，中国老年人的多维贫困指数整体下降，健康维度的贡献率上升。多维贫困模式可分为经济-精神贫困型、综合健康贫困型、能力-参与贫困型，女性、高龄、受教育程度低的老年人陷入各类贫困的风险均更高，农村、独居、家庭和社会支持匮乏的老年人更可能处于经济-精神贫困，与家人同住、社区健康服务和社会养老保障相对充分的老年人面临的综合健康贫困问题更突出，而缺乏家庭支持、社区休闲服务以及社会保障的老年人更可能陷入能力-参与贫困。

关键词: 老年贫困, AF方法, 多维贫困模式