bootstrapping is enough to test causal effects
Posted: Tue Apr 07, 2020 7:05 am
Dear,
Could you please explain how should I test causal effects in the following model:
(1) A, B, C, and D influence on X, Y.
(2) X, Y influence on M.
In bootstrapping, although the paths are significant, may I conclude that A, B, C, and D lead to X and Y, or that A, B, C, and D have associations with X and Y?
The results create a question of causality in the model; it is conceivable that M leads to X and Y, or that X and Y may lead to A, B, C, D? Should I do any steps to conclude that (1) A, B, C, and D lead to X, Y, and (2) X and Y lead to M?
Could you please explain how should I test causal effects in the following model:
(1) A, B, C, and D influence on X, Y.
(2) X, Y influence on M.
In bootstrapping, although the paths are significant, may I conclude that A, B, C, and D lead to X and Y, or that A, B, C, and D have associations with X and Y?
The results create a question of causality in the model; it is conceivable that M leads to X and Y, or that X and Y may lead to A, B, C, D? Should I do any steps to conclude that (1) A, B, C, and D lead to X, Y, and (2) X and Y lead to M?