Mediation Analysis Revisited Again

Australasian Marketing Journal - Tập 27 - Trang 52-56 - 2019
Felix Thoemmes1, Gunnar Lemmer2
1Department of Human Development, Cornell University, Martha Van Rensselaer Hall, Ithaca, NY 14853-4401, United States
2Psychologie, Philipps-Universität Marburg, G01 Institutsgebäude, 35032 Marburg, Germany

Tóm tắt

In a mediation model the effect of a dependent variable (DV) on an outcome is (partially) due to the DV's effect on one or multiple mediator(s) that consequently have an effect on the outcome. The use of such models as the theoretical background guiding empirical studies is widespread. Mediation models are fundamentally causal models that specify causal sequences. Unfortunately, the necessary causal assumptions are in practice often violated. In the current paper, we discuss possible improvements of causal mediation analyses, and highlight some potential pitfalls. We discuss the benefits gained by analyzing indirect effect between latent variables specified with measurement models. The validity of statistical findings can also be improved by using experimental designs. We discuss the cross-over design and the cross-over encouragement design, and how they can help improve causal conclusions. We also discuss recent advances on sensitivity analyses in the context of mediation models. Specifically, we explain how this analysis can be used to argue for the severity of unobserved confounding. Lastly, we discuss the practice of reversing the direction of the arrow between variables in a mediation model. We argue that if such reversals result in equivalent models, this practice cannot be recommended.

Tài liệu tham khảo

Baron R.M., 1986, J. Person. Soc. Psychol., 51, 1173, 10.1037/0022-3514.51.6.1173 Bollen K., 2002, Ann. Rev. Psychol., 53, 605, 10.1146/annurev.psych.53.100901.135239 Bullock J.G., 2010, J. Person. Soc. Psychol., 98, 550, 10.1037/a0018933 Giner-Sorolla R., 2016, J. Exp. Soc. Psychol., 65, 1, 10.1016/j.jesp.2016.01.010 Imai K., 2010, Psychol. Methods, 15, 309, 10.1037/a0020761 Imai K., 2010, Stat. Sci., 25, 51, 10.1214/10-STS321 Imai K., 2011, Am. Polit. Sci. Rev., 105, 765, 10.1017/S0003055411000414 Imai K., 2013, (with discussions) J. Royal Stat. Soc., Series A (Statistics in Society), 176, 5, 10.1111/j.1467-985X.2012.01032.x 10.1177/0193841X8100500502 10.1016/j.ausmj.2018.03.002 Lee S., 1990, Multivar. Behav. Res., 25, 313, 10.1207/s15327906mbr2503_4 Lemmer G., 2017, J. Exp. Soc. Psychol., 69, 144, 10.1016/j.jesp.2016.05.002 MacCallum R.C., 1993, Psychol. Bull., 114, 185, 10.1037/0033-2909.114.1.185 MacKinnon D. Introduction to Statistical Mediation Analysis2012 Routledge Stelzl I., 1986, Multiva. Behav. Res., 21, 309, 10.1207/s15327906mbr2103_3 Thoemmes F., 2015, Basic Appl. Soc. Psychol., 37, 226, 10.1080/01973533.2015.1049351 VanderWeele T.J., 2010, Epidemiology (Cambridge, Mass.), 21, 540, 10.1097/EDE.0b013e3181df191c VanderWeele T. Explanation in Causal Inference: Methods for Mediation and Interaction2015 Oxford University Press Verma T., 1990, Uncertainty in artificial intelligence 6, 220
Thông tin
Thông tin xuất bản

Australasian Marketing Journal

Tập 27

52-56

Thông tin tác giả