LGCM with latent variables PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Laura Johnson posted on Sunday, July 23, 2017 - 11:56 am
Hello, As part of a research study I am working on I am interested in looking at the factor structure of empowerment and then utilizing the latent empowerment variable to test for growth over four time points. However, I haven't found anything in the literature that talks about using an unobserved DV in a LGCM, all I have seen are observed. I wanted to see if this can be done (from a programmatic perspective but also in terms of whether this would be disadvantageous to run statistically)?

Thank you!
 Bengt O. Muthen posted on Sunday, July 23, 2017 - 5:29 pm
This is a standard approach. See Topic 4, slides 78 - from our short courses on our website.
 Laura Johnson posted on Tuesday, July 25, 2017 - 7:06 am
Great, thanks! The slides are helpful. This is a basic question but I want to check because I have four time points with six indicators per time point. When doing the EFA for each indicator at each time point, is the overall goal to have the same items in the indicator at all four time points? What do you do if they vary?

Thank you!
 Bengt O. Muthen posted on Tuesday, July 25, 2017 - 5:20 pm
The number of items can vary over the time points.
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