|
|
Fit indices with categorial and/or we... |
|
Message/Author |
|
|
dear drs. muthen, is there a way to obtain model fit indices like BIC when using CFA with the categorical and/or weight option? (as reported in muthen b., addiction 2006, "should substance use disorders be considered as categorical or dimensional?", table 1; binary items are used). is there a way to use the MIXED procedure to get that by suppressing latent classes? (because mixed when used for hybrid analysis reports these indices). thanks a lot in advance, michael höfler |
|
|
You can do this in two ways. You can use TYPE=MIXTURE; with one class, CLASSES= c(1); Or you can estimate the model using maximum likelihood, ESTIMATOR=MLR; instead of the default of weighted least squares. |
|
Back to top |
|
|