|
|
 |
| Using the 'standardize' option in mul... |
 
|
| Message/Author |
|
|
| Eunsoo Cho posted on Thursday, May 03, 2018 - 11:27 am
|
|
|
Drs. Muthen,
I analyzed data using the multiple group SEM. Despite running adequately with good model fit for measurement models with raw scores, structural models do not converge when we do not standardize some measures using its built-in option under Define command.
I understand that we lose variantion information by using the 'standardize' option but would it be okay if we build our structural model using standardized variables in the multigroup SEM? |
|
|
|
|
Don't use standardized variables for multiple-group analysis.
If you have very large variances for some variables, use Define to divide them (e.g. by 10).
If this doesn't help, send the non-converging output to Support along with your license number. |
|
|
| Back to top |
|
|
