Thank you, Gidon! very helpful hints!

I checked everything you recommended and all looked fine! I think I should rely on the results. How would you report the results? As always or any justifications for the high R-square?

Hi Benjamin,

It seems that you have some problem with the discriminant validity of your measures. The items of your measures may refer to the same construct that suggests items of the exogenous and endogenous variables are overlapping is first in my mind. More information about your model would be useful for more meaningful commentaries.

Hi Dmitry,

Thank you for the answer.

I've been thinking about the discriminant validity. I have three endogenous variables. Acceptance (5-items), Perceived Risk (2-items), and Perceived Benefit (5-items). They are somehow highly correlated with each other, however, the items used to measure them are conceptually different. The correlation is reasonable given the theory behind!

I had a similar problem with very high regression weights between exogenous and endogenous variables in SEM.  A journal editor said there must be a mistake but I couldn't find it.

Yes, in my experience, reviewers pay little attention to argumentation about theoretical reasonable links and may say these are problems due discriminant validity or multicollinearity for example. I think some visualization of the relationships would be useful to clarify details. You can start with applying scatter plot using factor scores to understand the relationships in your data.