Possibility of using Mplus for propos... PreviousNext
Mplus Discussion > Structural Equation Modeling >
 Sarah Rogers posted on Friday, October 04, 2013 - 1:40 pm
Dear Drs.Muthen

I'm currently exploring the possibility of using Mplus in my thesis to undertake SEM and wanted to check if my proposed analysis was possible before purchasing the software.

I'm looking at the relationship between family structure and child well-being.

1. My predictor variable is family type consisting of 3 categories - am I right in thinking I would create dummy variables to enter this into my model?

2. I have 3 mediating variables. One is an observed continuous variable (a measure of maternal mental health) the other two are latent variables ('parenting behaviours' and 'household economic circumstances')both of which consist of continuous, ordered categorical and binary observed indicators - can Mplus handle this mixture of variables in the context of latent mediating variables?

3. My outcome is a latent variable 'child well-being' which in turn consists of 4 latent variables representing 4 domains of child well-being. These 4 domains of child well-being consist of a mixture of continuous, ordered categorical and binary observed indicators. My understanding is that Mplus can handle this mixture of outcome variables?

I hope the explanation of my proposed model is clear enough and many thanks in advance for your help.
 Linda K. Muthen posted on Friday, October 04, 2013 - 4:30 pm
The answer to all of your questions is yes. Please note that latent variables are continuous irrespective of the scale of the factor indicators.
 Sarah  posted on Monday, October 07, 2013 - 8:12 am
Many thanks for your prompt reply.
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