Fixing a negative variance and outcomes PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Patchara Popaitoon posted on Sunday, November 20, 2011 - 4:45 am
Dear Linda,

I got a warning message re: negative redidual variance for an observed variable. This variable was detected. It was used in both indiviudal and group level analysis but the negative variance obtained when the variable was analyzed at the group level only.

The standardized and unstandardized residual variance for the variable are respectively,

-0.001 999.000 999.000 999.000; and
0.000 0.056 -0.002 0.999.

This residual variance was specified to zero in the next run (ACB by AOC1-AOC6; AOC6@0;) and this seems to solve the problem as there is no warning message reported.

The standardized and unstandardized residual variance obtained from the new analysis are respectively,
0.001 0.000 4.836 0.000; and
0.000 0.000 999.000 999.000.

However, the standardized factor score obtained for the variable is
1.000 0.000 13359.652 0.000.

I am not sure if this is commonly found when the residual variance is set to zero. Should I worry about it?

 Bengt O. Muthen posted on Sunday, November 20, 2011 - 8:01 pm
Please send your output to support.
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