The Importance of Covariate Selection in Controlling for Selection Bias in Observational Studies AUTHORS
نویسندگان
چکیده
The assumption of strongly ignorable treatment assignment is required for eliminating selection bias in observational studies. This article investigates how well this assumption is met by various conceptually unique sets of covariates. The methodology we use is a within-study comparison that contrasts results from a randomized experiment and a quasiexperiment that differ only in whether their respective treatment and comparison groups were formed at random or via self-selection. Where there was self-selection, various covariate sets and data-analytic methods were used to adjust for initial group selection differences and the adjusted effect sizes were then compared to the results of the experiment. In theory, the most important covariates for supporting strong ignorability are those that are closely related to both the real selection process and study outcomes. In the quasi-experiment, such variables reduced most of the bias due to self-selection whereas other covariates did not. The choice of ANCOVA or various propensity score methods played little role.
منابع مشابه
The importance of covariate selection in controlling for selection bias in observational studies.
The assumption of strongly ignorable treatment assignment is required for eliminating selection bias in observational studies. To meet this assumption, researchers often rely on a strategy of selecting covariates that they think will control for selection bias. Theory indicates that the most important covariates are those highly correlated with both the real selection process and the potential ...
متن کاملCausal Effect Identification by Adjustment under Confounding and Selection Biases
Controlling for selection and confounding biases are two of the most challenging problems in the empirical sciences as well as in artificial intelligence tasks. Covariate adjustment (or, Backdoor Adjustment) is the most pervasive technique used for controlling confounding bias, but the same is oblivious to issues of sampling selection. In this paper, we introduce a generalized version of covari...
متن کاملSelection Bias Correction in Supervised Learning with Importance Weight. (L'apprentissage des modèles graphiques probabilistes et la correction de biais sélection)
In the theory of supervised learning, the identical assumption, i.e. the training and the test samples are drawn from the same probability distribution, plays a crucial role. Unfortunately, this essential assumption is often violated in the presence of selection bias. Under such condition, the standard supervised learning frameworks may suffer a significant bias. In this thesis, we use the impo...
متن کاملSurvivor treatment selection bias in observational studies: examples from the AIDS literature.
Unlike patients in a randomized, clinical trial, patients in an observational study choose if and when to begin treatment. Patients who live longer have more opportunities to select treatment; those who die earlier may be untreated by default. These facts are the essence of an often overlooked bias, termed "survivor treatment selection bias," which can erroneously lead to the conclusion that an...
متن کاملUsing propensity score modeling to minimize the influence of confounding risks related to prenatal tobacco exposure.
INTRODUCTION Despite efforts to control for confounding variables using stringent sampling plans, selection bias typically exists in observational studies, resulting in unbalanced comparison groups. Ignoring selection bias can result in unreliable or misleading estimates of the causal effect. METHODS Generalized boosted models were used to estimate propensity scores from 42 confounding variab...
متن کامل