Omitted Variables, Countervailing Effects, and the Possibility of Overadjustment

The effect of conditioning on an additional covariate on confounding bias depends, in part, on covariates that are unobserved. We characterize the conditions under which the interaction between a covariate that is available for conditioning and one that is not can affect bias. When the confounding effects of two covariates, one of which is observed, are countervailing (in opposite directions), conditioning on the observed covariate can increase bias. We demonstrate this possibility analytically, and then show that these conditions are not rare in actual data. We also consider whether balance tests or sensitivity analysis can be used to justify the inclusion of an additional covariate. Our results indicate that neither provide protection against overadjustment

Conditioning · Confounding · Covariate · Econometrics · Omitted-variable bias · Statistics · Advanced Causal Inference Techniques · Mathematics · Statistical Methods and Bayesian Inference · Statistical Methods and Inference

• Control variables and causal inference

  Richard York•International Journal of Social…•2018

  Incluir item

• Methodological Uncertainty and Multi-Strategy Analysis

  Open Access•Laurence T Droy, Jarred Goodwin et al.•Bulletin of Sociological…•2020

  Incluir item

• Complementarity in alliances

  Open Access•J Andrés Gannon•American Journal of Political…•2025

  Incluir item

• Comparing Weighting Methods for Nonresponse in the HKPSSD Survey and the BCSPS

  T Cai, Hongyu Wang•Chinese Sociological Review•2018

  Incluir item

• We Ran 9 Billion Regressions

  Open Access•John Muñoz, Cristobal Young•Sociological Methodology•2018

  Incluir item

• Assessing Sensitivity to an Unobserved Binary Covariate in an Observational Study with Binary Outcome

  Open Access•Peter Rosenbaum, P R Rosenbaum et al.•Journal of the Royal Statistical…•1983

  Incluir item

• Causal Effects in Nonexperimental Studies

  Rajeev Dehejia, Rajeev H Dehejia et al.•Journal of the American…•1999

  Incluir item

• The central role of the propensity score in observational studies for causal effects

  Paul R Rosenbaum, Donald B Rubin•Biometrika•1983

  Incluir item

• How Important is Selection? Experimental VS. Non-Experimental Measures of the Income Gains from Migration

  Open Access•David Mckenzie, Steven Stillman et al.•Journal of the European Economic…•2010

  Incluir item

• Causal Diagrams for Treatment Effect Estimation with Application to Efficient Covariate Selection

  Halbert White, Xun Lu•The Review of Economics and…•2011

  Incluir item

• The Phantom Menace

  Open Access•Kevin A Clarke•Conflict Management and Peace…•2005

  Incluir item

• Return of the Phantom Menace

  Open Access•Kevin A Clarke•Conflict Management and Peace…•2009

  Incluir item

• A Cautionary Note on the Use of Matching to Estimate Causal Effects

  Open Access•Kevin Arceneaux, Alan S Gerber et al.•Sociological Methods & Research•2010

  Incluir item

• Comparing Experimental and Matching Methods Using a Large-Scale Voter Mobilization Experiment

  Open Access•Kevin Arceneaux, Alan S Gerber et al.•Political Analysis•2006

  Incluir item

• Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal Inference

  Open Access•Daniel E Ho, Katsushi Imai et al.•Political Analysis•2007

  Incluir item