For those who are familiar with multigroup analysis in AMOS:
- the general procedure is to constrain parameters and compare results by a chi-square (x2) difference. Ok. My question is relate to how we operationalize this. I have two options here:
a) first inform amos which are the groups, then fix the parameters, run the models and compare the x2 to the unconstrained model.
b) first fix the parameters and only then inform amos which are the groups and run the modes. By this way, amos understands that we are fixing parameters in both groups. By following this, when I free one of the constrained parameters I have to make it in both groups and run the model (i.e. I am comparing two models in which all parameters, except the one I want to test, are fixed in both groups.
Do you agree that this second procedure is the correct one?
Thanks in advance,
Try the all-new Yahoo! Mail . "The New Version is radically easier to use" The Wall Street Journal