Hello Witold,

the variance of V1 should be between 1.2 and 1.3 and from this variance 95.7%= 1.184 are left unexplained by the linear regression on V2. The other 4.3%= ? are negatively and obviously non-randomly related with V2. This assumes a simple linear regression without latent variables and no other observed variables in the model.

I hope this helps in starting with Mplus!

Hello Horst, 

On what basis do you know that the variance of V1 should be between 1.2 and 1.3? 

Thank you

In case of one predictor, you could calculate backward, if 95.7% of variance are unexplained ( = 1.184) , then 1.184/0.957 = total variance of your outcome, 1.237 in this case.