Analysis of change without latent var... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Luci M. Motoca posted on Wednesday, December 15, 2010 - 9:52 am
am interested to look at anxiety changes over time using a database with multiple time points. Anxiety was measured every other week while patients were in a treatment program. They completed a post treatment assessment assessment and were then asked follow up questions every other month until 1 year after the treatment at which point they completed a 1 year follow up assessment.

So, I have:

PRE Treatment week 2 week 4...week 12 (POST Tretament) and then month 2 month 4...month 12

I'm interested to see if there are different curves between pre to posttretament and how these curves look at follow up (week 12 to month 12 or pre to post using the different timepoints).

The problem that I have is that anxiety was assessed using 3 items at each time point, so I can't use latent variables. Is there a way to get around this limitation and still look at change in the way I described? Perhaps a different type of analysis other than latent growth curve modelling? Or can I use the 3 items to create the latent variable somehow? I'm thinking there must be a way to deal with this issue.

Thank you very much.
 Linda K. Muthen posted on Wednesday, December 15, 2010 - 5:20 pm
You could use a piecewise model to capture development before and after treatment.

Could you not use the three items as indicators of a factor of anxiety. See multiple indicator growth in the Topic 4 course handout on the website.
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