Abstract: There has been an agreement among international communities that social participation of the elderly people can actively cope with the issue of population ageing.However,most of the previous studies mainly focus on how social participation affecting health,little attention has been paid to the bi-directional relationship between the.Using the data of the Chinese Longitudinal Healthy Longevity Survey ( CLHLS) ,this paper explores this question using the logit and fixed effect models.After controlling for confounding factors,the model results show that there does exist significant bi-directional relationship between self-report health ( SRH) and social participation of the Chinese elderly.It is suggested that those older adults who report good SRH have a 22.0% ～ 40.1% higher odds ratio of engaging in social participation than those who have poor SRH; older adults who are engaged in social participation have a 16.4% ～ 25.6% higher odds ratio of reporting good SRH; the influence of SRH on social participation might be larger than that of social participation on SRH．Besides,the results reveal that the impact of social participation on SRH is more effective among elderly who report good SRH at baseline．

Keywords: Self-Reported Health, Social Participation, Bi-Directional Causal Relationship, Logit Model, Fixed Effect Model

摘要： 老年人社会参与已成为国际社会积极应对人口老龄化的普遍共识。以往研究多关注社会参与对老年人健康的影响,却忽视了二者之间可能存在的相互影响关系。文章采用中国高龄老人健康长寿跟踪调查(CLHLS)数据,利用Logit模型、固定效应模型分析和验证了二者之间的相互影响关系。在控制相关变量的前提下,模型表明,老年人自评健康与社会参与之间存在显著的互为因果关系,自评健康好的老年人比自评健康差的老年人进行社会参与的发生比高22.0%~40.1%;进行社会参与的老年人比无社会参与的老年人自评健康好的发生比高16.4%~25.6%;自评健康对社会参与的影响可能大于社会参与对自评健康的影响。研究还发现,社会参与在基期自评健康好的样本中更可能发挥显著影响。

关键词: 社会参与, 自评健康, 因果关系, Logit 模型, 固定效应模型