No CrossRef data available.
Article contents
Estimating controlled direct effects through marginal structural models
Published online by Cambridge University Press: 13 February 2020
Abstract
When working with panel data, many researchers wish to estimate the direct effects of time-varying factors on future outcomes. However, when a baseline treatment affects both the confounders of further stages of the treatment and the outcome, the estimation of controlled direct effects (CDEs) using traditional regression methods faces a bias trade-off between confounding bias and post-treatment control. Drawing on research from the field of epidemiology, in this article I present a marginal structural modeling (MSM) approach that allows scholars to generate unbiased estimates of CDEs. Further, I detail the characteristics and implementation of MSMs, compare the performance of this approach under different conditions, and discuss and assess practical challenges when conducting them. After presenting the method, I apply MSMs to estimate the effect of wealth in childhood on political participation, highlighting the improvement in terms of bias relative to traditional regression models. The analysis shows that MSMs improve our understanding of causal mechanisms especially when dealing with multi-categorical time-varying treatments and non-continuous outcomes.
- Type
- Original Articles
- Information
- Copyright
- Copyright © The European Political Science Association 2020
References
Acharya, A, Blackwell, M and Sen, M (2016) Explaining causal findings without bias: detecting and assessing direct effects. American Political Science Review 110, 512–29.CrossRefGoogle Scholar
Akee, R, Copeland, W, Costello, EJ, Holbein, JB and Simeonova, E (2018) Family income and the intergenerational transmission of voting behavior: evidence from an income intervention. NBER Working Paper 24770.CrossRefGoogle Scholar
Almond, G and Verba, S (1989) The Civic Culture: Political Attitudes and Democracy in Five Nations. Newbury Park, CA: Sage.Google Scholar
Beck, PA and Jennings, MK (1982) Pathways to participation. American Political Science Review 76, 94–108.CrossRefGoogle Scholar
Blackwell, M (2013) A framework for dynamic causal inference in political science. American Journal of Political Science 57, 504–520.CrossRefGoogle Scholar
Blackwell, M and Glynn, A (2014) How to Make Causal Inferences with Time-Series Cross-Sectional Data. Technical Report Working Paper. Harvard University.Google Scholar
Brady, HE, Verba, S and Schlozman, KL (1995) Beyond SES: a resource model of political participation. American Political Science Review 89, 271–294.CrossRefGoogle Scholar
Córdova, A (2009) Methodological note: measuring relative wealth using household asset indicators. Americas Barometer Insights 6, 1–9.Google Scholar
Currie, J (2008) Healthy, Wealthy, and Wise: Socioeconomic Status, Poor Health in Childhood, and Human Capital Development. Technical Report. National Bureau of Economic Research.CrossRefGoogle Scholar
Elwert, F and Winship, C (2014) Endogenous selection bias: the problem of conditioning on a collider variable. Annual Review of Sociology 40, 31–53.CrossRefGoogle ScholarPubMed
Glynn, AN and Quinn, KM (2010) An introduction to the augmented inverse propensity weighted estimator. Political Analysis 18, 36–56.CrossRefGoogle Scholar
Hernán, MÁ, Brumback, B and Robins, JM (2000) Marginal structural models to estimate the causal effect of zidovudine on the survival of HIV-positive men. Epidemiology 11, 561–570.CrossRefGoogle ScholarPubMed
Hernán, MÁ, Brumback, B and Robins, JM (2001) Marginal structural models to estimate the joint causal effect of nonrandomized treatments. Journal of the American Statistical Association 96, 440–448.CrossRefGoogle Scholar
Imai, K and Ratkovic, M (2014) Covariate balancing propensity score. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 76, 243–263.CrossRefGoogle Scholar
Imai, K and Ratkovic, M (2015) Robust estimation of inverse probability weights for marginal structural models. Journal of the American Statistical Association 110, 1013–1023.CrossRefGoogle Scholar
Imai, K, Keele, L and Tingley, D (2010) A general approach to causal mediation analysis. Psychological Methods 15, 309–334.CrossRefGoogle ScholarPubMed
Imai, K, Keele, L, Tingley, D and Yamamoto, T (2011) Unpacking the black box of causality: learning about causal mechanisms from experimental and observational studies. American Political Science Review 105, 765–789.CrossRefGoogle Scholar
Jennings, MK, Markus, GB, Niemi, RG and Stoker, L (2005) Youth-parent socialization panel study, 1965–1997: four waves combined.CrossRefGoogle Scholar
Kang, JDY and Schafer, JL (2007) Demystifying double robustness: a comparison of alternative strategies for estimating a population mean from incomplete data. Statistical Science 22, 523–539.CrossRefGoogle Scholar
Lipset, SM (1959) Some social requisites of democracy: economic development and political legitimacy. American Political Science Review 53, 69–105.CrossRefGoogle Scholar
Miller, AH, Gurin, P, Gurin, G and Malanchuk, O (1981) Group consciousness and political participation. American Journal of Political Science 25, 494–511.CrossRefGoogle Scholar
Miratrix, LW, Sekhon, JS and Yu, B (2013) Adjusting treatment effect estimates by post-stratification in randomized experiments. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 75, 369–396.CrossRefGoogle Scholar
Montgomery, JM and Olivella, S (2018) Tree-based models for political science data. American Journal of political Science 62(3), 729–744.CrossRefGoogle Scholar
Montgomery, JM, Nyhan, B and Torres, M (2018) How conditioning on posttreatment variables can ruin your experiment and what to do about it. American Journal of Political Science 62, 760–775.CrossRefGoogle Scholar
Moore, JC and Welniak, EJ (2000) Income measurement error in surveys: a review. Journal of Official Statistics 16, 331.Google Scholar
Mutz, DC (2002) The consequences of cross-cutting networks for political participation. American Journal of Political Science 46, 838–855.CrossRefGoogle Scholar
Nandi, A, Glymour, MM, Kawachi, I and VanderWeele, TJ (2012) Using marginal structural models to estimate the direct effect of adverse childhood social conditions on onset of heart disease, diabetes, and stroke. Epidemiology 23, 223–232.CrossRefGoogle Scholar
Ojeda, C (2018) The two income-participation gaps. American Journal of Political Science 62, 813–829.CrossRefGoogle Scholar
Pearl, J (2001) Direct and indirect effects. Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence (UAI '01). San Francisco, CA: Morgan Kaufmann Publishers Inc., pp. 411–420.Google Scholar
Pearl, J (2011) The Mediation Formula: A Guide to the Assessment of Causal Pathways in Nonlinear Models. Technical Report DTIC Document.CrossRefGoogle Scholar
Platt, RW, Delaney, JAC and Suissa, S (2012) The positivity assumption and marginal structural models: the example of warfarin use and risk of bleeding. European Journal of Epidemiology 27, 77–83.CrossRefGoogle ScholarPubMed
Robins, JM (1997) Causal inference from complex longitudinal data. In Berkane, M. (eds) Latent Variable Modeling and Applications to Causality. New York, NY: Springer, pp. 69–117.CrossRefGoogle Scholar
Robins, JM (1999) Association, causation, and marginal structural models. Synthese 121, 151–179.CrossRefGoogle Scholar
Robins, JM, Hernán, MÁ and Brumback, B (2000) Marginal structural models and causal inference in epidemiology. Epidemiology 11, 550–560.CrossRefGoogle ScholarPubMed
Rosenbaum, PR (1984) The consequences of adjustment for a concomitant variable that has been affected by the treatment. Journal of the Royal Statistical Society. Series A (General) 147, 656–666.CrossRefGoogle Scholar
Rubin, DB (1977) Assignment to treatment group on the basis of a covariate. Journal of Educational and Behavioral Statistics 2, 1–26.CrossRefGoogle Scholar
Samii, C, Paler, L and Daly, S (2017) Retrospective causal inference with machine learning ensembles: an application to Anti-Recidivism Policies in Colombia. Political Analysis 24, 434–456.CrossRefGoogle Scholar
VanderWeele, TJ (2009) Marginal structural models for the estimation of direct and indirect effects. Epidemiology 20, 18–26.CrossRefGoogle ScholarPubMed
VanderWeele, TJ (2010) Bias formulas for sensitivity analysis for direct and indirect effects. Epidemiology 21, 540–551.CrossRefGoogle ScholarPubMed
Verba, S and Nie, NH (1972) Participation in America: Political Democracy and Social Equality. Chicago, IL: University of Chicago Press.Google Scholar
Verba, S, Nie, NH and Kim, J (1978) Participation and Political Equality: A Seven-Nation Comparison. Chicago, IL: University of Chicago Press.Google Scholar
Watkins, S, Jonsson-Funk, M, Brookhart, MA, Rosenberg, SA, O'shea, TM and Daniels, J (2013) An empirical comparison of tree-based methods for propensity score estimation. Health Services Research 48, 1798–1817.Google ScholarPubMed
Westreich, D, Cole, SR, Schisterman, EF and Platt, RW (2012) A simulation study of finite-sample properties of marginal structural Cox proportional hazards models. Statistics in Medicine 31, 2098–2109.CrossRefGoogle ScholarPubMed
Torres Supplementary Materials
Torres Supplementary Materials
PDF
224.5 KB