Heckman selection model - suitable for categorical and continuous DVs?

Hello everyone,

I am trying to figure out whether Heckman selection can be and should be used for my data. I have the same group of investors who are first of all asked about their willingness to invest in a specific project (on a 1-7 scale) and afterwards, after being presented with additional information, they are asked whether they will actually invest (0-1 variable). I want to explore whether there is a different set of predictors for individuals' reported willingness to invest, versus whether they actually invest. Can I use the Heckman selection in this case, such that in the first equation I have data from all participants (willingness), and in the second I censor it on whether they actually invested or not?

I am using stata's heckman twostep procedure but not sure if I am allowed to apply this model given my data.

Your help will be greatly appreciated!