Control group in Latent Class Analysis PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Nic Kim posted on Thursday, November 12, 2009 - 4:31 am
Dear, Dr. Muthen.

I am a beginner who is working on a project using LCA. A few guidance as to where to start to look at would be greatly appreciated.

I have continuous variables that are collected during a series of small-group activities. Small numbers of people were randomly assigned to groups to interact with their group members.

All groups (so,all subjects) engaged in the same kind of activity. One of the questions of the research was how one's behavior is affected by other group members' behavior.

As an exercise, I tried a series of LCA on the whole data and it worked well. But I think I should control group effect because people from the same latent class could have behaved differently based on the overall behavioral dynamics occurred in their assigned experimental groups. Or people from different latent class could have behaved similarly.

Is my concern a valid one?

I think I need to do something similar to group dummy or covariate in regression analysis but I am not sure.

I searched the discussion pages and user's guide but without sufficient background knowledge, it is hard to know where to begin.

What types of analysis am I looking for and what types of command option are likely to be relevant to my question?

Thank you.

 Linda K. Muthen posted on Thursday, November 12, 2009 - 3:05 pm
I think you want to regress the classes on a covariate as is shown in Example 7.12.
 Nic Kim posted on Friday, November 13, 2009 - 6:39 am
Thank you, Dr.Muthen.

I will look into that.
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