|
Message/Author |
|
|
I am trying to specify a model in which there are two DV's, both of which are indicated by three binary variables. I have tried setting up the model the same way as I would with DVs indicated by variables that are continuous and the model ran, but I got a non positive definite message. I was told that the error of the DV's needs to be set to zero, but I am not sure exactly how to do that. I'm wondering what I need to do differently in setting up a model with DVs indicated by binary variables instead of continuous variables. |
|
|
There is no difference in setting up the model for continuous or categorical factor indicators. However, with categorical factor indicators, residual variances are not parameters in the model so they cannot be fixed to zero. You would need to change the model. |
|
Back to top |
|
|