|
|
 |
| Indirect effects in Bayesian multigro... |
 
|
| Message/Author |
|
|
|
|
I am trying to assess structural invariance of a mediation model across four groups using a Bayesian approach, but cannot seem to get it to work for the indirect effect. Here is the syntax I have been trying, for which Mplus keeps telling me that "ind1" is an unknown parameter label in the MODEL PRIORS command:
Model: %overall%
M on X (a#);
Y on M (b#);
Y on X (c#);
Model Constraint:
ind1 = a1*b1;
ind2 = a2*b2;
ind3 = a3*b3;
ind4 = a4*b4;
Model Priors:
DIFF(ind1 ind2 ind3 ind4) ~ N(0, 0.01) |
|
|
|
|
| Model Priors can refer to parameter labels given in the Model command, not Model Constraint quantities. |
|
|
| Back to top |
|
|
