因果推断五十年:成就、挑战与应对

被引:1
作者
许琪
机构
[1] 南京大学社会学院
关键词
因果推断; 鲁宾因果模型; 观察研究; 因果效应; 因果机制;
D O I
暂无
中图分类号
C3 [社会科学研究方法];
学科分类号
摘要
鲁宾因果模型问世五十年以来,学术界对因果关系的理论认识和分析方法有了长足发展,应用因果推断方法开展的实证研究也不断增多。但鲁宾因果模型在取得巨大成就的同时,也面临很多挑战和局限。这既包括将该模型应用于观察研究时无法保证随机分配假定的局限,也包括为了满足近似随机分配的合理性,而无视研究问题的实质重要性的局限。除此之外,鲁宾因果模型在分析结果的原因和识别因果机制方面也存在不足。为了应对这些挑战,研究者既要改进定量研究设计,也要有意识地借鉴和吸纳定性分析工具,只有将定性和定量两种方法有机结合起来,才能对因果关系进行全面和透彻的分析,进而有助于对因果问题的全面认识,促进定量与定性两个研究范式的融通和促进。
引用
收藏
页码:131 / 144
页数:14
相关论文
共 34 条
[1]  
Identification of Causal Effects Using Instrumental Variables.[J].Joshua D. Angrist;Guido W. Imbens;Donald B. Rubin.Journal of the American Statistical Association.2012, 434
[2]   Comparing IV with structural models: What simple IV can and cannot identify [J].
Heckman, James J. ;
Urzua, Sergio .
JOURNAL OF ECONOMETRICS, 2010, 156 (01) :27-37
[3]   The Credibility Revolution in Empirical Economics: How Better Research Design is Taking the Con out of Econometrics [J].
Angrist, Joshua D. ;
Pischke, Joern-Steffen .
JOURNAL OF ECONOMIC PERSPECTIVES, 2010, 24 (02) :3-30
[4]   FOR OBJECTIVE CAUSAL INFERENCE, DESIGN TRUMPS ANALYSIS [J].
Rubin, Donald B. .
ANNALS OF APPLIED STATISTICS, 2008, 2 (03) :808-840
[5]   Improving causal inference - Strengths and limitations of natural experiments [J].
Dunning, Thad .
POLITICAL RESEARCH QUARTERLY, 2008, 61 (02) :282-293
[6]   Experimental study of inequality and unpredictability in an artificial cultural market [J].
Salganik, MJ ;
Dodds, PS ;
Watts, DJ .
SCIENCE, 2006, 311 (5762) :854-856
[7]   Doubly robust estimation in missing data and causal inference models [J].
Bang, H .
BIOMETRICS, 2005, 61 (04) :962-972
[9]   Using Maimonides' rule to estimate the effect of class size on scholastic achievement [J].
Angrist, JD ;
Lavy, V .
QUARTERLY JOURNAL OF ECONOMICS, 1999, 114 (02) :533-575
[10]   Matching as an econometric evaluation estimator: Evidence from evaluating a job training programme [J].
Heckman, JJ ;
Ichimura, H ;
Todd, PE .
REVIEW OF ECONOMIC STUDIES, 1997, 64 (04) :605-654