It was about percentages, and quite a big difference so it's not just because he was in the news more. But you are entirely correct that it didn't differentiate in levels of negativity or positivity, but that's true for most studies I found on this. I think it's almost impossible to have an entirely objective research on this because selling what is neutral is inherently subjective. A politician could simply get more negative reporting because he says more incorrect things or has flawed plans, or he could by the victim of bias.
Fair enough.
Of course there's no study about all media but you don't have to. If you include the big ones, they cover a vast majority of media reach to the public, and the ones you left out can never tip the balance because they are too small.