×

A simple method for purging mediation effects. (English) Zbl 1437.62294

Summary: Mediation effects occur when an independent variable lies in the causal path of another independent variable, thereby swamping its effects. Mediation effects can impact the ability to accurately estimate some quantity of interest. A simple solution for addressing this issue is regressing the treatment variable on the mediator variable and then use the stored residuals from this bivariate specification as the new “direct” variable. This residualized purging process, which is the core of a recently released R package, purging, is detailed in this paper. The goal at present is to offer researchers and practitioners an application using both simulated and real data to introduce and demonstrate the value of the purging solution, applied via the purging package, to efficiently address mediation effects. A second order goal of the paper is to underscore the value of writing software to streamline research processes.

MSC:

62J12 Generalized linear models (logistic models)
62-08 Computational methods for problems pertaining to statistics
62P25 Applications of statistics to social sciences

Software:

R; medeff; Stata; purging
PDF BibTeX XML Cite
Full Text: DOI

References:

[1] Baron, RM; Kenny, DA, The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations, J Personal Soc Psychol, 51, 6, 1173 (1986)
[2] Frisch, R.; Waugh, FV, Partial time regressions as compared with individual trends, Econom J Econom Soc, 1, 4, 387-401 (1933) · JFM 59.1207.14
[3] Hicks, R.; Tingley, D., Causal mediation analysis, Stata J, 11, 4, 605 (2011)
[4] Huang, FL, Using instrumental variable estimation to evaluate randomized experiments with imperfect compliance, Pract Assess Res Eval, 23, 2, 2 (2018)
[5] Imai, K.; Keele, L.; Yamamoto, T., Identification, inference and sensitivity analysis for causal mediation effects, Statist Sci, 25, 1, 51-71 (2018) · Zbl 1328.62478
[6] Imai K, Keele L, Tingley D, Yamamoto T (2010) Causal mediation analysis using R. In: Advances in social science research using R. Springer, New York, NY, pp 129-154
[7] Imai, K.; Keele, L.; Tingley, D.; Yamamoto, T., Unpacking the black box of causality: learning about causal mechanisms from experimental and observational studies, Am Polit Sci Rev, 105, 4, 765-789 (2011)
[8] Long, JS, Regression models for limited and categorical dependent variables (1997), Thousand Oaks, CA: Sage Publications Inc, Thousand Oaks, CA
[9] Lovell, MC, A simple proof of the FWL theorem, J Econ Educ, 39, 1, 88-91 (2008)
[10] Rucker, DD; Preacher, KJ; Tormala, ZL; Petty, RE, Mediation analysis in social psychology: current practices and new recommendations, Soc Pers Psychol Compass, 5, 6, 359-371 (2011)
[11] UCLA: Statistical Consulting Group (2018) Hsbdemo Data Set. https://stats.idre.ucla.edu/stat/data/hsbdemo.dta. Accessed 13 Apr 2018
[12] Waggoner, Philip D., Do constituents influence issue-specific bill sponsorship?, Am Polit Res, 47, 4, 709-738 (2019)
[13] Zhao, X.; Lynch, JG Jr; Chen, Q., Reconsidering Baron and Kenny: myths and truths about mediation analysis, J Consum Res, 37, 2, 197-206 (2010)
[14] Zhou, Xiang; Wodtke, Geoffrey T., A regression-with-residuals method for estimating controlled direct effects, Polit Anal, 27, 3, 360-369 (2019)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.