I am estimating the transactional associations of two LGC processes (maternal depressive symptoms and child externalizing behaviors) and would like to include the concurrent covariances of their observations in my estimations (i.e. wave 1 MDS with wave 1 CEB). However, I am receiving the following error message from Mplus:

*** ERROR in MODEL command

Covariances for censored, count or nominal variables with other observed variables

are not allowed.

I looked at Mplus' disccusion board for answers, but the only ones I could find are related to how residual variances among indicators of a latent variables can be accounted for. I could not find any information regarding accounting for the covariance of two indicators, belonging to separate processes/latent variables.

Mplus support is not available until June 10th, and I wanted to see if anyone else came across a similar complication, and found a solution.

Thank you in advance for any help and/or ideas!