摘要
流动人口的健康选择性一直是学界争论的焦点,有的研究认为,健康选择性会随着人口流动的普遍化而消失,这是值得商榷的。因此,文章利用中国家庭追踪调查(Chinese Family Panel Studies,CFPS)2010和2012年两轮数据,先从截面数据和跟踪样本两个角度,结合Logistic回归方法,讨论流动人口的健康选择性;再针对跟踪样本中的新增流动人口,结合适合稀少事件的firthlogit回归和倾向得分匹配方法,以自评健康为实验变量(treatment),在剔除自评健康的内生性后,从反事实因果推论的平均实验效应角度,讨论并检验了流动人口健康选择性的存在。结论认为,流动人口健康选择性依然存在;之所以在一般的统计模型中不显著的原因可能来自于自评健康的内生性和各种特征对健康选择性的替代作用。
Using two waves national-representation data - Chinese Family Panel Studies （CFPS） , this paper examine the ＇healthy migrant hypothesis＇ in a the social context of contemporary China. The Logistic regression models are used to the cross sectional and followed samples at first. Then, an appropriate comparison is used between the new migrant during 2010 and 2012 and the non-migrants at two waves among the followed sample, while propensity score matching methods is used to eliminate the selection bias of the self-rate health, which is regarded as the treatment, based on the counterfactual causal inference framework. Consistent with research from both international migration contexts and other internal migration settings, this paper finds that migrants are positively selected on the basis of health, and the relationship between health and migration will not diminishes across time. The non-significant relationship in regression models maybe caused by the endogeneity of sell;rate health and the exchangeable replacement of other migration selection characteristics to self-rate health.
作者
王伶鑫
周皓
WANG Ling-xin;ZHOU Hao(Department of Sociology,Tsinghua University,Beijing 100084,China;Department of Sociology,Peking University,Beijing 100871,China)
出处
《西北人口》
CSSCI
2018年第6期13-22,31,共11页
Northwest Population Journal
基金
北京市流动人口的社会融合研究(编号:13SHA001)
关键词
流动人口
健康
自评健康
选择性
内生性
Migration
Health
Self-rate health
Selection
Endogeneity
