Regression-based mediation analysis: a formula for the bias due to an unobserved precursor variable

다국어 초록 (Multilingual Abstract)

Researchers want to know whether the change in an explanatory variable X affects the change in a response variable Y (i.e., X causes Y). In practice, there can be two causal paths from X to Y, the path through a mediating variable M (indirect effect) and the path not through M (direct effect). The parameter estimation and hypothesis testing can be performed by a regression-based mediation model. It is already known that randomization of X is not enough for unbiased estimation, and the bias due to an unobserved variable has been discussed in literature but often overlooked. In this article, we first review the challenge under a simple mediation model, then we provide a formula for the exact bias due to an unobserved precursor variable W, the variable which potentially causes the changes in X, M, and/or Y. We present simulation studies to demonstrate the impact of an unobserved precursor variable on hypothesis testing for indirect effect and direct effect. The simulation results show that the inflation of type I error is serious particularly in a large sample study. To numerically demonstrate the formula of the exact bias, a popular data set published in a journal of statistics education is revisited, and we quantify why the conclusion of data analysis can be different before and after accounting for the precursor variable. The result shall remind the importance of a precursor variable in mediation analysis.

참고문헌 (Reference)

1 Hong, G., "Weighting-based sensitivity analysis in causal mediation studies" 43 (43): 32-56, 2018

2 Rohrer, J. M., "Thinking clearly about correlations and causation : Graphical causal models for observational data" 1 (1): 27-42, 2018

3 Baron, R. M., "The moderator-mediator variable distinction in social psychological research : Conceptual, strategic, and statistical considerations" 51 (51): 1173-1182, 1986

4 Emery, C., "The importance of being psychologically empowered : Buffering the negative effects of employee perceptions of Leader–Member Exchange differentiation" 92 (92): 566-592, 2019

5 Fisher, R. A., "Statistical methods for research workers" Oliver and Boyd 1925

6 Garcia, P. R. J. M., "Role modeling as a socialization mechanism in the transmission of career adaptability across generations" 111 : 39-48, 2018

7 Hong, G., "Ratio-of-mediator-probability weighting for causal mediation analysis in the presence of treatment-by-mediator interaction" 40 (40): 307-340, 2015

8 Hall, N. S., "R. A. Fisher and his advocacy of randomization" 40 : 295-325, 2007

9 Caniëls, M. C., "Proactivity and supervisor support in creative process engagement" 37 (37): 188-197, 2019

10 Manuti, A., "People make the difference : An explorative study on the relationship between organizational practices, employees’ resources, and organizational behavior enhancing the psychology of sustainability and sustainable development" 11 (11): 1499-, 2019

11 Villaluz, V., "Ownership and leadership in building an innovation culture" 40 (40): 138-150, 2018

12 Kang, J., "Overview and practice of causal inference in observational studies" 1 (1): 00002-, 2014

13 Hayes, A. F., "Methodology in the social sciences. Introduction to mediation, moderation, and conditional process analysis: A regression-based approach" Guilford Press 2013

14 Seli, P., "Increasing participant motivation reduces rates of intentional and unintentional mind wandering" 83 (83): 1057-1069, 2017

15 Imai, K., "Identification, inference and sensitivity analysis for causal mediation effects" 25 : 51-71, 2010

16 Guber, D. L., "Getting what you pay for: The debate over equity in public school expenditures" 7 (7): 1999

17 Zhang, L., "Exploring the interpersonal determinants of job embeddedness and voluntary turnover : A conservation of resources perspective" 29 (29): 413-432, 2019

18 VanderWeele, T. J., "Explanation in causal inference: Methods for mediation and interaction" Oxford University Press 2015

19 Zhu, H., "Does relationship matter? How social distance influences perceptions of responsibility on anthropomorphized environmental objects and conservation intentions" 95 : 62-70, 2019

20 Rachel C. Adams, "Claims of causality in health news: a randomised trial" Springer Science and Business Media LLC 17 (17): 91-, 2019

21 Glass, T. A., "Causal inference in public health" 34 : 61-75, 2013

22 VanderWeele, T. J., "Bias formulas for sensitivity analysis for direct and indirect effects" 21 (21): 540-551, 2010

23 Hayes, A. F., "Beyond Baron and Kenny : Statistical mediation analysis in the new millennium" 76 (76): 408-420, 2009

24 Preacher, K. J., "Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models" 40 : 879-891, 2008

25 Hair J. F., "A primer on partial least squares structural equation modeling (PLS SEM)" 2014

26 Imai, K., "A general approach to causal mediation analysis" 15 (15): 309-334, 2010