Broad topics we might want to look at

I think there's a dangerous confounding variable, insofar as news networks draw viewers that share ideological inclination (the Republican-inclined choose to watch Fox), and moreover adjust their stances to match what they think their viewers want or to fill a perceived niche in the larger news market (as MSNBC may currently be attempting to do). If we're gathering cross-sectional data (as we will be, yes?), I'm hard-pressed to think of a way to unravel that, given that respondents will be poor at judging to what extent their viewpoints led to their choice of news, and to what extent they have evolved from their news choices.

Not to be unconstructive, it's a really good question to ask, I'm just drawing a blank at present on how we might home in on it...