|
Message/Author |
|
|
Hello, I'm working on writing a grant and need to say something about multi-group modeling. I of course know of chi-square difference testing and using WLSMV that the difference is not distributed as a chi-square. However, I need to say something about what we are doing. When the user's guide states that the derivatives of the less-restricted H1 model are saved for the second test, what are these derivaties? In other words, they are the derivatives of...? PS, I know what derivatives are but am not sure how they are used in the chi-square difference testing. Thanks,DR |
|
|
It is the derivatives that are taken with respect to the parameters of the model. They are used to create the estimated information matrix as described in Muthen (1984) and Muthen and Satorra (1985). |
|
|
Thank you DR |
|
Back to top |
|
|