社会科学中的因果推断与异质性偏差
CAUSAL INFERENCE AND HETEROGENEITY BIAS IN SOCIAL SCIENCE.
作者信息
Xie Yu
机构信息
University of Michigan.
出版信息
Inf Knowl Syst Manage. 2011 Jan 1;10(1):279-289. doi: 10.3233/IKS-2012-0197.
Abstract
Because of population heterogeneity, causal inference with observational data in social science may suffer from two possible sources of bias: (1) bias in unobserved pretreatment factors affecting the outcome even without treatment; and (2)bias due to heterogeneity in treatment effects. Even when we control for observed covariates, these two biases may occur if the classic ignorability assumption is untrue. In cases where the ignorability assumption is true, "composition bias" can occur if treatment propensity is systematically associated with heterogeneous treatment effects.
摘要
由于人口异质性，社会科学中利用观测数据进行因果推断可能会受到两种潜在偏差来源的影响：（1）即使在未接受治疗的情况下，未观察到的预处理因素对结果产生的偏差；（2）治疗效果异质性导致的偏差。即使我们控制了观测到的协变量，如果经典的可忽略性假设不成立，这两种偏差仍可能出现。在可忽略性假设成立的情况下，如果治疗倾向与异质的治疗效果存在系统性关联，就可能出现“构成偏差”。
相似文献
1
CAUSAL INFERENCE AND HETEROGENEITY BIAS IN SOCIAL SCIENCE.社会科学中的因果推断与异质性偏差
Inf Knowl Syst Manage. 2011 Jan 1;10(1):279-289. doi: 10.3233/IKS-2012-0197.
2
Population heterogeneity and causal inference.人群异质性与因果推断。
Proc Natl Acad Sci U S A. 2013 Apr 16;110(16):6262-8. doi: 10.1073/pnas.1303102110. Epub 2013 Mar 25.
3
Propensity Score-Based Methods versus MTE-Based Methods in Causal Inference: Identification, Estimation, and Application.因果推断中基于倾向得分的方法与基于边际治疗效应的方法：识别、估计与应用
Sociol Methods Res. 2016 Feb;45(1):3-40. doi: 10.1177/0049124114555199. Epub 2014 Nov 3.
4
Causal inference accounting for unobserved confounding after outcome regression and doubly robust estimation.结果回归和双重稳健估计后考虑未观察到的混杂因素的因果推断。
Biometrics. 2019 Jun;75(2):506-515. doi: 10.1111/biom.13001. Epub 2019 Mar 29.
5
Folic acid supplementation and malaria susceptibility and severity among people taking antifolate antimalarial drugs in endemic areas.在流行地区，服用抗叶酸抗疟药物的人群中，叶酸补充剂与疟疾易感性和严重程度的关系。
Cochrane Database Syst Rev. 2022 Feb 1;2(2022):CD014217. doi: 10.1002/14651858.CD014217.
6
Practice of causal inference with the propensity of being zero or one: assessing the effect of arbitrary cutoffs of propensity scores.基于倾向值为零或一的因果推断实践：评估倾向得分任意截断值的效果
Commun Stat Appl Methods. 2016 Jan;23(1):1-20. doi: 10.5351/CSAM.2016.23.1.001. Epub 2016 Jan 31.
7
The control outcome calibration approach for causal inference with unobserved confounding.有未观测混杂时因果推断的控制结局校准方法。
Am J Epidemiol. 2014 Mar 1;179(5):633-40. doi: 10.1093/aje/kwt303. Epub 2013 Dec 20.
8
Estimating Heterogeneous Treatment Effects with Observational Data.利用观测数据估计异质性治疗效果。
Sociol Methodol. 2012 Aug;42(1):314-347. doi: 10.1177/0081175012452652.
9
Inverse probability weighting with error-prone covariates.带有易出错协变量的逆概率加权法。
Biometrika. 2013;100(3):671-680. doi: 10.1093/biomet/ast022.
10
The Mechanics of Omitted Variable Bias: Bias Amplification and Cancellation of Offsetting Biases.遗漏变量偏差的机制：偏差放大与抵消偏差的消除
J Causal Inference. 2016 Sep;4(2). doi: 10.1515/jci-2016-0009. Epub 2016 Nov 8.
引用本文的文献
1
Heterogeneous Effects in Education: The Promise and Challenge of Incorporating Intersectionality Into Quantitative Methodological Approaches.教育中的异质性效应：将交叉性纳入定量方法论的前景与挑战。
Rev Res Educ. 2018 Mar;42(1):72-92. doi: 10.3102/0091732x18759040. Epub 2018 Apr 5.
2
Weak signal detection: A discrete window of opportunity for achieving 'Vision 90:90:90'?弱信号检测：实现“90-90-90”目标的一个短暂机会窗口？
SAHARA J. 2016;13(1):17-34. doi: 10.1080/17290376.2015.1123642.
3
Feeling good about the iron rice bowl: Economic sector and happiness in post-reform urban China.对“铁饭碗”的良好感受：中国改革开放后城市中的经济部门与幸福感
Soc Sci Res. 2015 Sep;53:203-17. doi: 10.1016/j.ssresearch.2015.05.008. Epub 2015 May 29.
4
Estimating the heterogeneous relationship between peer drinking and youth alcohol consumption in Chile using propensity score stratification.使用倾向得分分层法估计智利同龄人饮酒与青少年酒精消费之间的异质性关系。
Int J Environ Res Public Health. 2014 Nov 17;11(11):11879-97. doi: 10.3390/ijerph111111879.
本文引用的文献
1
Estimating Marginal Returns to Education.估算教育的边际回报。
Am Econ Rev. 2011 Oct;101(6):2754-2781. doi: 10.1257/aer.101.6.2754.
2
Estimating Heterogeneous Treatment Effects with Observational Data.利用观测数据估计异质性治疗效果。
Sociol Methodol. 2012 Aug;42(1):314-347. doi: 10.1177/0081175012452652.
3
Heterogeneity in returns to college education: selection bias in contemporary taiwan.大学教育回报的异质性：当代台湾的选择性偏差。
Soc Sci Res. 2011 May;40(3):796-810. doi: 10.1016/j.ssresearch.2010.12.008.
4
Smoking and lung cancer: recent evidence and a discussion of some questions.吸烟与肺癌：近期证据及若干问题探讨
J Natl Cancer Inst. 1959 Jan;22(1):173-203.
5
Heterogeneity's ruses: some surprising effects of selection on population dynamics.异质性的策略：选择对种群动态的一些惊人影响。
Am Stat. 1985 Aug;39(3):176-85.
6
The philosophical foundations of Darwinism.达尔文主义的哲学基础。
Proc Am Philos Soc. 2001 Dec;145(4):488-95.
7
Estimating causal effects from large data sets using propensity scores.使用倾向得分从大型数据集中估计因果效应。
Ann Intern Med. 1997 Oct 15;127(8 Pt 2):757-63. doi: 10.7326/0003-4819-127-8_part_2-199710151-00064.
Zotero 插件