1) For a continuous X, the two X values (exposure levels) that the effects com- pare are given in parentheses after X in the IND statement:
.... X(x1 x0);
where you can choose x0 as say the mean of X and x1 say 1 SD above the mean.
2) Linear effects because Y is continuous.
We ask that no more than one window is used per posting. If longer descriptions are needed, please send to Support.
jan mod posted on Friday, February 10, 2017 - 12:39 pm
Thank you very much and sorry for the double post. I have three more questions.
1) X(x1 x0): does Mplus then use the exposure variable as binary for the indirect effect? So section 8.4.1 (on a binary mediator is then equivalent for continuous X with that statement. 2) PNDE is here then interpreted as the linear effect of x1 on y with the mediator that is allowed to vary like it would if those respondents would have been exposed to x0. 3) my TNIE and PNIE are the same. Is this possible?
1) No, the exposure variable is kept as continuous. It is just that your effect is evaluated as a function of a change in X from x0 to x1. See Section 2.6.1 of our book for an explanation of the case of a continuous exposure (although in the case of a continuous mediator and moderation but that doesn't matter).
3) In the general case they are not the same but perhaps without an interaction (moderation) they could be - I haven't worked through the formulas for this but see the formulas of 8.4.1.