Abstract

this paper is to bring to the attention of investigators several basic limitations of the associational criterion. We will show that the associational criterion does not ensure unbiased e#ect estimates, nor does it follow from the requirement of unbiasedness. After demonstrating, by examples, the absence of logical connections between the statistical and the causal notions of confounding, we will de#ne a stronger notion of unbiasedness, called stable unbiasedness, relative to which a modi#ed statistical criterion will be shown necessary and su#cient. The necessary part will then yield a practical test for stable unbiasedness which, remarkably, does not require knowledge of all potential confounders in a problem. Finally,wewill argue that the prevailing practice of substituting statistical criteria for the e#ect-based de#nition of confounding is not entirely misguided, because stable unbiasedness is in fact what investigators have been and should be aiming to achieve, and stable unbiasedness is what statistical criteria can test.

Citations

5662	Probabilistic reasoning in intelligent systems: networks of plausible inference - Pearl - 1988
249	Estimating Causal Effects of Treatments in Randomized and Nonrandomized Studies - Rubin - 1974
175	TS: A theory of inferred causation - Pearl, Verma
113	JM: A new approach to causal inference in mortality studies with sustained exposure periods – Application to control of the healthy worker survivor effect. Mathematical Modeling - Robins - 1986
81	KJ Modern Epidemiology - Rothman - 1986
78	Graphical models, causality, and intervention - Pearl - 1993
75	Planning of Experiments - Cox - 1958
35	Case-control studies. Design, conduct, analysis - Schlesselman - 1982
34	Causal diagrams for experimental research. Biometrika 82(4):669–710 - Pearl - 1995
29	The analysis of case-control studies - BRESLOW, DAY - 1980
26	JM: Identifiability, exchangeability, and epidemiological confounding - Greenland, Robins - 1986
20	Vector Autoregressions for Causal Inference - Leamer - 1985
14	CR: Toward a clearer definition of confounding - Weinberg - 1993
13	Estimating causal e ects of treatments in randomized and nonrandomized studies - Rubin - 1974
10	Confounding: essence and detection - Miettinen, Cook - 1981
9	Adjusting for covariates that have the same distribution in exposed and unexposed cohorts - Gail - 1986
8	Effects of collapsing multidimensional contingency tables - Bishop - 1971
8	Collapsibility of relative risk in contingency tables with a response variable. JR Statist Soc B. 1992;54:585–93. Publish with Libertas Academica and every scientist working in your field can read your article “I would like to say that this is the most au - Geng
7	Confounding confounding - Grayson - 1987
7	Identi ability, exchangeability, and epidemiological confounding - Greenland, JM - 1986
6	Book reviews - Imbens - 1997
4	TR: Confounding in epidemiologic studies: the adequacy of the control group as a measure of confounding. Biometrics - Wickramaratne, Holford - 1987
3	A consequence of omitted covariates when estimating odds ratios - Hauck, Heuhaus, et al. - 1991
2	Re: ``Confounding confounding - Greenland, Morgenstern, et al. - 1989
2	Epidemiologic Research. Lifetime Learning - Kleinbaum, Kupper, et al. - 1982
2	Causal effects of dynamic policies - Robins, Pearl - 1996
2	Individual causal effects, average causal effects, and unfoundedness in regression models - Steyer, Gabler, et al. - 1996
2	Designing case-contol studies. Environmental health perspectives - Yanagawa - 1984
1	Reiersol, Geary and the idea of instrumental variables - Aldrich - 1993
1	Causal diagrams for epidimiologic research - Greenland, Pearl, et al. - 1997
1	Modern epidemiology. Brown Little, 1st edition - Rothman - 1986
1	Collapsibility of multide - Whittemore - 1978
1	Causal e ects of dynamic policies - Robins, Pearl - 1997
1	Individual causal e ects, average causal e ects, and unconfoundedness in regression models - Steyer, Gabler, et al. - 1996
1	Toward a clearer de nition of confounding - Weinberg - 1993