Strange model result PreviousNext
Mplus Discussion > Structural Equation Modeling >
 ellen posted on Tuesday, May 22, 2012 - 5:26 am
Hi Dr. Muthen,

I encountered a strange result in a SEM model. I have 5 latent variables in total. Two latent variables ("racism" and "self-stigma") are not correlated in the measurement model. But in the structural model, "racism" significantly predicted "self-stigma" in negative direction (opposite to what would be expected).

My question is: How can two latent variables be unrelated, but then have a significant path coefficient between them in a structural model?

Below is what I cut and paste from the output:

RACISM WITH (estimate) (S.E.) (Est./S.E.) p-value
SELFSTGMA 0.144 0.085 1.698 0.089

SELFSTGMA ON (estimate) (S.E.) (Est./S.E.) p-value
RACISM -0.307 0.145 -2.121 0.034

Could you please let me know what could be the problem here?

Thanks so much!
 Linda K. Muthen posted on Tuesday, May 22, 2012 - 11:40 am
Please send the relevant outputs and your license number to
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message