|
|
| Message/Author |
|
|
|
|
Hello,
I am trying to remove unwanted correlations between factors and I'm having some difficulty. Here is the code I'm working with below. I don't want RunExp to be correlated with my latent variables (NE, INVHED, INVCENT, etc.). I've tried a few different methods with no success. Also, if I attempt to delete too many arrows in the diagrammer it will freeze. Thanks!
model:
NE by NE1 NE2 NE3;
INVHED by INVHED1 INVHED2 INVHED3 INVHED4;
INVCENT BY INVCENT1 INVCENT2 INVCENT3 INVCENT4 INVCENT5;
INVSOC BY INVSOC1 INVSOC2 INVSOC3;
INVSELF BY INVSELF1 INVSELF2 INVSELF3;
INVSOID BY INVSOID1 INVSOID2 INVSOID3;
CHAIR BY CHAIR1 CHAIR2 CHAIR3 CHAIR4;
INVHED on NE;
INVCENT on NE;
INVSOC on NE;
INVSELF on NE;
INVSOID on NE;
CHAIR on NE;
RunExp on NE;
Avghrs on NE;
FamSup on NE;
INVHED3 with INVHED1;
INVHED4 with INVHED2;
INVSOC1 with INVSOC3;
INVCENT3 with INVCENT2;
INVCENT4 with INVCENT2;
INVSELF3 with INVSELF1 |
|
|
|
|
| runexp WITH ne@0; fixes the covariance between runexp and ne to zero. |
|
|
| Back to top |
|
|
