Invariance test on a mediation model
I am running an invariance test on a mediation model using SEM, to check whether the model is invariant between genders and across countries. I've selected gender as a grouping variable and ticked loadings and intercepts for equality constraints. From the output, estimates were only constrained for the mediator and outcome variable but not for predictors (still differed between genders). In the means table, std.errors were zero for both predictors and there were also no z-value and p-value for them. I am not sure what I might have done wrong here. Could someone provide any suggestion please?
Thanks a lot,