Help needed with variance error PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 Elpida posted on Sunday, August 25, 2019 - 7:58 pm
Am currently trying to conduct an LPA and I am unsure why an error keeps appearing after entering my input.

The output says:

*** WARNING in MODEL command
All variables are uncorrelated with all other variables within class.
Check that this is what is intended.
*** ERROR
One or more variables have a variance greater than the maximum allowed of 1000000.
Check your data and format statement or rescale the variable(s) using the DEFINE command.

Here are the variances of the variables that mPlus produced in the output:

Continuous Number of
Variable Observations Variance

**ID 1228 **********
BORDER 1228 0.255
AGE 1217 271.314
SEX 1228 0.194
HBDCD 1195 6049.204
HBDID 1195 2183.282
STNOAPD 1168 14282.457
STTCD 1140 15095.090
STTID 1140 8083.069
MTNCD 1192 4721.972

These variances were cross checked with SPSS and were found to be consistent.

Could someone please kindly let me know why this error is happening when the variables are correlated and the variance is lower than 1000000 for all variables?

Thank you
 Bengt O. Muthen posted on Monday, August 26, 2019 - 6:48 am
It looks like you use your ID variable among your analysis variables. Either you don't have a USEV list or you have the ID variable on your USEV list.

It sounds like you are using mixture (latent class) modeling. The warning refers to correlations within a class. That doesn't say that your variables are uncorrelated. A latent class variable describes correlations among your observed variables but there is no correlation within class.
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: