|
|
 |
| exact same degrees of freedom |
 
|
| Message/Author |
|
|
| Yiran Su posted on Friday, July 20, 2018 - 7:32 pm
|
|
|
Hi All,
I am wondering how the structural model and measurement model have the exact same degrees of freedom? |
|
|
|
|
That could be because the structural part of your model has no restrictions on it. For instance, if your measurement model has 3 correlated factors and your structural model regresses one of the factors on the other two, the two models have the same df because you have the same number of parameters for your factor covariance matrix as for your structural model (6 in both cases) and there is a one-to-one translation between those two sets of parameters.
If this doesn't help, send your output to Support along with your license number. |
|
|
| LaylaS posted on Thursday, July 26, 2018 - 8:53 am
|
|
|
Thank you for your reply!
Based on the model, I think should not have the same degree of freedom. Because I have four latent variables for CFA and four latent variables which have 5 paths for structural model. But the CFA model and SEM model have exactly the same fit index. However, using this code with another year's data, I did not have this issue.
MODEL:
PHYEN by PhyEnv_2 PhyEnv_3 PhyEnv_4;
SERVE by Qual_5 Qual_3 Qual_4;
ATMOS by Atmos1 Atmos2 Atmos3;
CONTRI by Contri2 Contri3 Contri4;
COG by Cogimg2 Cogimg3 Cogimg4 Cogimg6 Cogimg7;
SUBBRA by PHYEN SERVE ATMOS;
BRF by BrandF1 BrandF2 BrandF3;
COG on BRF SUBBRA CONTRI;
CONTRI on BRF SUBBRA; |
|
|
|
|
| Send the 2 outputs you refer to to Mplus Support along with your license number. |
|
|
| Back to top |
|
|
