Message/Author |
|
|
Hello I have completed invariance testing which tells me that I have non-invariance at the scalar level. This is fine and expected. However, I'm not sure how to find out which intercepts are causing the non-invariance. I have read in various places about gradually restricting the intercepts one at a time to find where the misfit is coming from. Is there a certain part of the user guide you can refer me to, with specific instructions on how to do this? Or another source with instructions? Thanks. |
|
|
Use Modindices in your scalar run to see which equality constraint should be relaxed. |
|
|
Thank you. For further clarification - What does it mean if the MI is high in one group, but not the other? For example: I have one factor that is 10.457 in one group, but 29.599 in the other group (and this would be the largest MI in the analysis). But the next highest factor is 17.964 in both groups. So should I relax the first factor or the second factor? I really appreciate your help! |
|
|
Please send your output to Support along with your license number and point to the part of the output you are referring to. |
|
|
Hi again, I think I worked out the issue. When I removed the [intercepts] from the syntax, I had left the [factor means@0] in there. And that's why I wasn't able to understand my output. Am I correct in understanding that both of these need to be removed at the same time to do intercept invariance correctly? Can you explain why that is? (I'm just curious as to the method behind it) |
|
|
I don't know your exact input, but mentioning the intercepts in a particular group frees it up so the default invariance no longer holds. |
|
Back to top |