I have a problem with r-square. URGENT¡

[gemmafu]

Hello:

I have a problem, when I calculate the explained variance of each of the latent variables that I have in my model, the sum of all the variances explained does not match the r squared of the endogenous variable.
What can happen?
I have mediator variables, but the total effects I have controlled for these variables and match those in the smartpls 3 report. But multiplying the path coefficient by the correlation of latent variables and then adding them does not match r squared.
Can you help me?
Thank you.

[cringle]

Your constructs probably are not orthogonal and, thus, your observation makes total sense.

Kind regards
Christian