| Message/Author |
|
|
|
|
Hello,
I'm working with a scale with negatively and positively worded items. The theory suggests that its factor solution should be unidimensional with correlated errors of the NW items.
The scale is a twelve version item. lets say var1 to var12. Also im using WLSMV.
How can I include in the Mplus model this by using CTCU?
thanks in advance, |
|
|
|
|
| Correlated errors are handled by WITH statements also for WLSMV. |
|
|
|
|
Thanks for the answer Dr. Muthen.
I stated the correlated errors by using the WITH statement but it didīt work. The employed instruction is as follow
Model:
Factor by var1 var2 var3 var4 var5 var6 var7;
var1 WITH var2 var 1 WITH var3 and so on...
Is this the correct employment of the command?
Thanks |
|
|
|
|
You have to say:
var1 WITH var2; var 1 WITH var3; etc |
|
|
| Back to top |
