I think one of the problems with citing that first study as evidence Russian disinfo is targeted at conservatives more than liberals is that it only studied one case, and Russian disinformation campaigns tailor their disinfo to different demographics, often through brute force/trial and error. So it is quite possible that the particular case they studied happens to be tailored to (or more successfully resonated with) conservatives, while another specific case would have resonated with liberals more thus resulting in more liberal exposure by their metrics.