|
|
 |
| Type = complex with multiple imputations |
 
|
| Message/Author |
|
|
|
|
Dear Dr. Muthen,
I am using version 8.0 with both add-ons, and am attempting to run a cross-lagged model with clustered data using type=complex, estimator= wlsmv, and imputations for missing data. My colleague was able to run this model on a Mac but for some reason I keep getting this error message:
*** ERROR in ANALYSIS command
TYPE=COMPLEX is not currently available with multiple imputation with the
DATA IMPUTATION command.
Do you have any idea what is causing this?
Thank you. |
|
|
|
|
It is is caused by the fact that you are trying to do the imputation and the estimation in one step - which is not available for type=complex. Also Mplus can't do it in one step because it doesn't know how to use the clustering - it would require to specify a two level model. The proper way to accomplish this is in two steps.
Step 1. Multilevel data imputation: for example take User's guide example 11.8. If you don't want to specify an imputation model you can run instead an H1 imputation as follows: remove the model statement and use this
ANALYSIS: TYPE = TWOLEVEL BASIC;
ESTIMATOR = ml;
Step 2. Analyze the data - see user's guide example 12.6 STEP 2 (but replace type=montecarlo with type=imputation) |
|
|
| Back to top |
|
|
