|
|
 |
| Using latent covariance matrix as input |
 
|
| Message/Author |
|
|
|
|
Greetings,
Let's say that I have a first-order measurement model for which I want to "explore" the higher-order structure (i.e. ESEM), which is not directly possible.
I could export the latent covariance matrix from the initial measurement model to a new data set, and use it as the data for the higher-order analysis.
My understanding is that I can get the latent mean and covariance matrix with the following command:
SAVEDATA:
TECH4 IS tech4.dat;
But then, I'm not sure how exactly to have the program read this new data file (not sure the format of the tech4.dat file).
Is it:
DATA:
File is tech4.dat;
TYPE is FULLCOV MEANS;
or
TYPE is MEANS FULLCOV ;
or
TYPE is COVARIANCE MEANS;
or
TYPE is MEANS COVARIANCE;
And then I guess the variable order corresponds to their order of appearance int the initial input ?
Many thanks |
|
|
|
|
| I believe either of the last two TYPE options will work with TECH4. You can see the order by looking at the SAVEDATA summary at the end of the output where TECH4 is saved. |
|
|
| Back to top |
|
|
