How to test correlation PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 yuming ning posted on Thursday, November 13, 2008 - 10:25 am
i have a simple question for coding
my hypothesized model is
A->D->E
B->D->E
A->E
B->E
C->E
hence my codes will be
E on A B C D;
D on A B;
but how do i test A<->B (ie, cov(A,B))?
i tried adding (A with B;) into above codes, but the result gave me estimates for all of cov(A,B), cov(A,C), and cov(B,C). i only want cov(A,B) and assume
cov(A,C)=cov(B,C)=0
 Linda K. Muthen posted on Thursday, November 13, 2008 - 11:00 am
In regression, the means, variances, and covariances of the covariates are not parameters in the model. The model is estimated conditioned on the covariates. You should not mention a WITH b. If you want to know the correlation between a and b, see the sample statistics.
 Anna  posted on Tuesday, June 15, 2010 - 7:08 am
I have a question concerning the test of correlations. How do I test if correlations between two latent factors (e.g. between level and slope) are significantly different between groups or time points? Is this possible in Mplus?

Thank you very much for your help!
 Linda K. Muthen posted on Tuesday, June 15, 2010 - 9:24 am
You can test if the covariances between two growth factors are the same for two groups using either MODEL TEST of chi-square or loglikelihood difference testing.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: