I am trying to test a complex hybrid model (with a second order latent factor measurement component) which performed well for my overall dataset across two groups. MPlus runs the same exact model across all data, with no identification problems detected (although two of the second order latent factors have two latent indicators each).
As soon as I add my grouping variable, and at the outset of testing for fit of the common measurement model, I get the following error:
"THE MODEL ESTIMATION TERMINATED NORMALLY
THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES COULD NOT BE COMPUTED. THE MODEL MAY NOT BE IDENTIFIED. CHECK YOUR MODEL. PROBLEM INVOLVING PARAMETER 294."
Any tips on how to resolve this error without having to change the theoretically grounded model?