Missing data on covariates PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Sun Kim posted on Thursday, April 05, 2012 - 2:23 am
Dear Dr. Muthen,

I am trying to fit an SEM with 10 covariates I wish to control for.
As you suggested in previous threads, I specified the covariate variances in the model command so that they don't end up being dropped with list-wise deletion.

Here is a part of my model (the whole model is much more complex):

MODEL:

achieve by lw02 pc02 prob02;

abilityM by ab1m ab2m ab3m ab4m ab5m ab6m ab8m ab9m ab10m;

abilityR by ab1r ab2r ab3r ab4r ab5r ab6r ab8r ab9r ab10r;


childsex; age02; black; white; hispanic; famincome; edD02; edM02; bioP; achieve97;

achieve on childsex age02 black white hispanic famincome edD02 edM02 bioP achieve97;

abilityR on childsex age02 black white hispanic famincome edD02 edM02 bioP achieve97;
abilityM on childsex age02 black white hispanic famincome edD02 edM02 bioP achieve97;


The above model does not converge.
However, I noticed that this model does not have any problems fitting when I get rid of the categorical covariates (childsex, black, white, hispanic, bioP).
Even when I add the categorical covariates in the regression and do not even model their variances, the model does not converge.

I would like to include categorical and continuous covariates and not have them list-wise deleted.
Do you have any suggestions?
Thank you.
 Linda K. Muthen posted on Thursday, April 05, 2012 - 12:10 pm
If you are putting the categorical covariates on the CATEGORICAL list, remove them. This is for dependent variables only. Otherwise, send the output and your license number to support@statmodel.com.
 Sun Kim posted on Thursday, April 05, 2012 - 4:02 pm
I'm sorry where can I find my license number?
 Linda K. Muthen posted on Thursday, April 05, 2012 - 4:39 pm
You can login to your account to see your license number. If you have problems doing this, contact Mplusadmin@statmodel.com.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: