* what is the source? do they have a reputable history?
* do they have conflicts of interest?
* do they have a history of faithfully printing retractions and admitting mistakes, or a history of exaggerating facts?
* do they have a history of writing accurately about subjects I am an expert about? (e.g. subfields of computer science)
If they score pretty well, I tend to believe in the facts (and evaluate the interpretation) they espouse until I come across something that contradicts it. So far the only source I regularly trust is the Economist, but I haven't researched the history of many news organizations. It's just a starting point for me.
That's just a somewhat extreme illustration of how you, and everybody else, has to rely on heuristics to evaluate statements. It's impossible to even prove a tenth of percentage points of the facts you rely on daily from first principle.