|
|
Multilevel Multiple Group Analysis |
|
Message/Author |
|
|
Dear Mplus team, I'm running a MGA to establish invariance of a ML model of 177 individuals nested in 34 teams, and the two groups are US and Europe - the model doesn't run, and I get the below message: THE MODEL ESTIMATION DID NOT TERMINATE NORMALLY DUE TO AN ILL-CONDITIONED FISHER INFORMATION MATRIX. CHANGE YOUR MODEL AND/OR STARTING VALUES. I have already tried setting the very low between-level variances@0, but no success. The error message continues telling me: THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES COULD NOT BE COMPUTED. Any advice would be much appreciated. |
|
|
As a first step,I would try each group separately. If you continue to have problems, send the appropriate files and your license number to support@statmodel.com. |
|
|
Thank you Prof Muthen. This particular model runs for the European group separately (24 teams, 129 individuals), but NOT for the US group separately (10 teams, 48 individuals). (With a modified models using item parceling, it is the other way around) |
|
|
I don't know how many parameters your model has, but 48 individuals in 10 clusters is quite small. It is recommended to have a minimum of 30 clusters. Note also that if the same model does not fit well in each group, comparisons across groups are not valid. |
|
Back to top |
|
|