Multivariate multilevel model:grouped... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 R McDowell posted on Friday, July 29, 2016 - 2:23 am
Hello. I wish to find out if it is possible to fit the model below in Mplus.

Outcome 1 is a proportion, measured over time e.g. 3/4, 6/7, 0/3 etc.

Outcome 2 is also a proportion measured at the same time points, but the numerators differ from Outcome 1, eg. 2/3, 0/5 etc.

Both outcomes have been measured in the same population and are not necessarily independent.

I can model each outcome separately (outcome as a function of time), using individual binary data eg for outcome one the dependent variable in the first four records will comprise of 3 zeros and 1 one etc. As there is a cluster level variable, it isn't possible to use the freqweight option to collapse the data. However I would like to model both outcomes simultaneously. As both outcomes have been measured from the same sample, are not necessarily independent, I would prefer not to use a multigroup approach. It is possible to set up the data for this type of multivariate analysis in Mplus?

Many thanks for any advice anyone can offer.
 Bengt O. Muthen posted on Friday, July 29, 2016 - 3:10 pm
Sounds like your outcomes are really counts. Bivariate modeling of counts is possible adding a factor that influences both outcomes.
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