So who exactly who do you trust?
Trust is earned, and can be lost (harder to earn back if you do that, even more so if maliciously rather than incompetence). You can also have to do your own analysis to tease things out.
Those that are impartial are best but they can be harder to find so you also get to do analysis of motives, incentives (incentives make for outcomes so very often) and interests on top of things. Equally incentives can include those that I would not normally trust, or trust in other scenarios, as it is in their best interest to tell the truth -- the old legal adage of "When the facts are on your side, pound the facts. When the law is on your side, pound the law. When neither is on you side, pound the table" working in politics as well so getting your facts on point and unimpeachable such that they have to try the other two becomes a tacit admission in at least western societies (communist where everybody lies about everything, albeit with very good incentives to do so, and the idea of face in Asia can make things a bit more fun there).
I like numbers (especially things that can be backtested and thus used as predictive models) but always be wary of who is making them and grouping them for you, I like revealed preferences (people can say a lot but actions louder than words), I like secondary indicators* that most lying are less inclined (or in the case the all too common incompetence then unable) to consider (for inflation, assuming we care about prices rather than money supply for this one, then producer price index and energy costs would be a leading indicator and not one those creating consumer price indexes, which are super selective, care for).
I also find most people have domains of interest and knowledge that are rather specific to them, even more so in combination with other things. Worthiness of a general source can then come from how those that would not necessarily know those fields plays out.
*the various investment firms buying in satellite photos to count cars in the car park as a predictor of shopper count. One example of that sort of thing.
Can be more difficult than simply accepting whatever the news might say or government drones (elected and otherwise) might say.
This all rolls into my general thesis on such things in that everybody is self interested, possibly even sociopathic on occasion**, cunt that works to said own interest with as much foresight as the maths provides (quarterly or annual in business, next election in politics) and as much insight as can be expected (given the complexity of the world then that is not all that much, makes my fondness for numbers occasionally a bit harder to justify, though that is the reason for diversification of your portfolio and cycling things out before they get too tainted). Play to that and assume people are going to fuck it up before too terribly long, usually by overextending themselves with short term goals at the cost of long term, and not much will surprise you and your predictions will not be far wrong.
**sociopaths can work in politics and business (cold risk assessment and no personal loyalty is good stuff to have, though can come with a downside when predicting psychology of the masses), as can "normal" people/followers, the so called saints that make up the other percentage of that aspect of psychology can not thrive in the slightest in such scenarios and get ground down and out. Leave them to play medic, certain charities (though even those mostly go from above), your nice maintenance men that keep your expenses down a bit by going above and beyond, possibly teachers (though lots of questions there too, not to mention most of that falls from on high anyway) and such like.