As promised, here is my Food Pyramid of News:

This pyramid is restricted to news sources, not information sources in general. A balanced diet of information would include books, (more) journal articles, and possibly online coursework like MRUniversity.
Blogs being at the bottom will be controversial to some people, but I stand by it. The blogosphere is an extremely large, rich, and diverse place. If I had to read news in only one medium for the rest of my life, I would choose blogs, hands down, no question. Unlike radio and newspapers/magazines, it is conversational (between blogs) and interactive (between bloggers & commenters), which makes an enormous difference in terms of how much you can learn about the world there. Note that if a blog is hosted on the website of a major newspaper (like WonkBlog or Paul Krugman’s blog) then I count it as a blog, not a newspaper, just for the purposes of this pyramid (i.e. for the purpose of deciding how much time to spend reading which types of news media). Whether a blog is on a major newspaper’s website or not is sort of irrelevant otherwise of course.
Some people will want to switch Newspapers/Magazines with Radio/Podcasts, and in fact I’m one of those people, but I put them the way I did though because that’s what I think is more natural for more people. It depends on how you spend your time, I think; people who spend a lot of time commuting, travelling, and exercising are going to find podcasts and radio a more convenient way to consume the kind of news you get from radio, newspapers, and magazines. It doesn’t make a huge difference though. The point is just that you need sources of boilerplate, just-the-facts versions of various news stories. The home pages of major newspapers are good for these, as is NPR. But don’t too much time there; diversify!
Social Media items can be blog posts, newspaper articles, or podcasts, discovered via Facebook, Twitter, Reddit, or other sites. The point is that some non-trivial amount of time should be spent finding items in those media via social media instead of via the homepages of major newspapers, blogs etc., because of how social media uncovers “unknown unknowns,” i.e. items you otherwise would not have discovered had you not been aware of the source before you found it via social media. I put this in a small box though, only because it can be difficult to separate the wheat from the chaffe on social media sites. The wheat, though, is still worth separating.
“Primary Sources,” for lack of a better term for a catch-all category, includes things like Fed data, journal articles, industry reports, think-tank publications, and anything else that might be cited in a newspaper article or blog post and summarized second-hand. Very often a blogger or journalist will report using a primary source in the public domain, but s/he does so in an arbitrary, biased, or misleading way.
For example, a lot of people cite Card & Krueger’s famous minimum wage study, and interpret it to mean that “empirically, the minimum wage has no effect on unemployment.” But without reading the paper yourself, or at least perusing it, you’ll be left wondering whether it really does mean what they say it means (or, worse, you’ll just accept the journalist’s or blogger’s interpretation as fact). It’s worth spending the time to read it for yourself, figuring out what the methodology was, becoming literate in the study itself rather than just the interpretation, seeing where/when/how the study was done, putting it in the context of a larger literature, discovering that literature for yourself, and evaluating the findings for yourself, on the basis of your own reading of that study. That is a relatively labor-intensive example, but it doesn’t need to be. Things like inflation rates are much easier to fact-check.
So, visit those primary sources if they are publicly available, and don’t be afraid to delve into something you don’t understand very well. It’s worth spending as much time doing this as you spend on Twitter and Facebook, no doubt.
A lot of people with advanced degrees are going to spend a lot more time checking out things like FRED data, journal articles, and think-tank pubs, but even for these folks I still think this pyramid works. The reason is that much of what I’ve called “Primary Sources” here is not “news” to these people, but their profession. Again, this pyramid is really just about news consumption, not information consumption generally, and as far as news consumption is concerned, that box works even for folks with advanced degrees.
Finally, television occupies a tiny little triangle at the very top. Don’t watch cable, don’t watch Fox News, nor CNN, nor MSNBC, or (almost) anything else on television, for the sake of news. Television is for entertainment, not news. Do try to get a sense of what certain major television pundits’ views are on hot-button issues, but not because these people know what they’re talking about, but because a large number of Americans take those views seriously and it’s good to know what many Americans think. Personally I do not take these people seriously at all, at least as far as their shows are concerned (there are some pundits, like Bill O’Reilly and Rachel Maddow, whose books I take a little more seriously than their shows). This triangle is basically the “sugar/sweets” section, i.e. mostly for entertainment, not news. Other than keeping a pulse on those television pundits views, I don’t watch any news shows other than The Daily Show and The Colbert Report, which, again, are purely for entertainment purposes.
This pyramid isn’t going to reflect everybody’s preferences, but I think it’s a good starting point for a lot of people. If I had to guess how most people currently consume their news, I would guess that they basically flip everything upside down, with television on bottom and blogs near the top. Primary sources are probably in the top triangle, which is tragic.
I’d appreciate constructive feedback in the comments section. What does your pyramid look like? I’m especially interested to know what news pyramids look like for people with different interests from myself. This pyramid works well for an economics/politics person; what about an evolutionary biologist?