And what would that show, exactly? That donors are more likely to donate to politicians who agree with them? In order to be interesting, you'd have to demonstr

And what would that show, exactly? That donors are more likely to donate to politicians who agree with them? In order to be interesting, you'd have to demonstrate that a donation caused the politician's stance and was not just because of the politician's stance - and I'm not sure that's extractable from the data set you've got to work with.