I have a concern using a probit model. In short, I am using cross-sectional data to investigate the impact of digitalisation and it's various tools on the propensity to adopt various sustainable behaviours for European SME's. As most of my dependent variables are binary, I ran several probit models to analyze this relationship (using Stata). My concern is the potential reverse causality of my models.

I would therefore very much appreciate if someone could enlighten me on this problematic. How can I test for reverse causality in probit models using Stata? If there is indeed a justified concern for reverse causality, is there another model that allows me to investigate a hand in hand relationship rather that a causal relationship?

Any help or information related to these topics is very welcome.

Best regards