本文利用2012年中国家庭追踪调查（CFPS）数据，在考虑选择性偏差和异质性的基础上，运用倾向分值匹配的方法，分别估计了高中教育和高等教育对健康的影响。利用局部线性匹配得到的结果证实了更高的受教育水平可以产生更好的健康水平；利用有序Logistic 模型比较匹配前后样本的回归系数，证明了选择性偏差的存在；异质性分析结果表明，高中教育对健康的影响存在正异质性，高等教育对健康的影响存在负异质性。%The latest data from China Family Panel Studies (2012) is employed to investigate the causal effect of education on health for high school graduates and college graduates respectively. Taking selective bias and heterogeneity into account, we choose propensity score matching approach, and local linear matching estimation is the method for matching process. The results indicate that higher education generates more health benefits; selective bias is detected by ordered-logistic regression coefficients. Heterogeneous effect of education on health is positive for high school graduates, but negative for college degree-holders.
展开▼
