. Author manuscript; available in PMC: 2021 Feb 2.

Published in final edited form as: Struct Equ Modeling. 2020 Aug 3;27(6):975–984. doi: 10.1080/10705511.2020.1777133

Table 1.

Regression-based estimators of the causal indirect, direct, and total effects for models with a continuous mediator and a continuous outcome.

Effect	Potential outcomes notation	Causal estimator
TNIE	E[Y_i(1, M_i(1)) − Y_i(1, M_i(0))]	ba + ha
PNIE	E[Y_i(0, M_i(1)) − Y_i(0, M_i(0))]	ba
TNDE	E[Y_i(1, M_i(1)) − Y_i(0, M_i(1))]	c’ + hi₁+ ha
PNDE	E[Y_i(1, M_i(0)) − Y_i(0, M_i(0))]	c’ + hi₁
CDE	E[Y_i(1, m) − Y_i(0, m)]	c’ + hm
TE	E[Y_i(1, M_i(1)) − Y_i(0, M_i(0))]	c’ + hi₁ + ba + ha

Abbreviations: TNIE, total natural indirect effect; PNIE, pure natural indirect effect; TNDE, total natural direct effect; PNDE, pure natural direct effect; CDE, controlled direct effect; TE, total effect.