Mplus Discussion > Multilevel Data/Complex Sample >
 David Hussey posted on Wednesday, November 03, 2010 - 3:58 am
I wish to model survey data for school pupils collected in 50 (sampled) schools, with two classes (sampled) in each school (one in each of two years). There was no sampling within classes. Please can you confirm that the most appropriate modeling framework for this data would be TYPE=COMPLEX TWOLEVEL.

Also, I would like to check my understanding of WITHIN and BETWEEN. I wish to model two independent varaiables (A and B say) where A varies within class and B is measured at the school-level. Clearly A should be specified as WITHIN but what about B given that it is measured at the school level and not the class level?

Finally, is there any way to run many models at once in MPlus or automate a process using some kind of macro? I wish to run the same model with about 100 different dependent variables.

Thank you for your help.
 Linda K. Muthen posted on Wednesday, November 03, 2010 - 12:33 pm
If you put the school-level variables on the BETWEEN list and use them on the between level, the standard errors will be correct.

See if the information at the following link might be helpful about running analyses with different dependent variables:

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