Help with CFA on longitudinal data PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
 Martin Zwatschek posted on Tuesday, August 07, 2012 - 5:26 am
Hi all,

I am working on my diploma thesis and was advised to use MPlus to do a CFA on my longitudinal data. After quite some literature on the topic, I didnīt make much progress, though. And now hope anyone here can help me, maybe give me a syntax to go with and explaing what it is that I have to do. (sorry for being such a newbie, but this kind of statistics is quite new for me).

I have two variables from my data (say X and Y).
Each variable was queried 14 times over a week (morning and evening). So I have X1m through X7m, and X1e through X7e and the same for Y.
With those I want to confirm a little modell which consists of 2 factors (A and B) where A comprises all the morning variables (1m through 7m) and B all the evening variables (1e through 7e).

I read about structural equation models (LS(T) models) and Multilevel models. First of all I donīt know which one to use and second, I donīt know how the syntax should look like for either one. The examples in the book donīt give much away for just this case.

Can anyone help me out here. Cause Iīm blank.

Thanks so much.
 Linda K. Muthen posted on Tuesday, August 07, 2012 - 2:36 pm
The Topic 4 course handout and video on the website covers multiple indicator growth starting at Slide 78. This shows how to test for measurement invariance over time and how to estimate a growth model for the factors.

You might find a wider range of help on a general discussion forum like SEMNET.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message