I thought it would make sense for a first post on a writing blog to be a "how to" exploration. The "how" is "research" and the "to" is "to create a more believable story that people will actually want to read."
It amazed me as I got more into writing that so many authors, when asked how they did it--how they came up with characters, or plots, worlds, and so on--rarely, if ever, mention research. The usual answer to that question is always something glib, "If I told you I'd lose my job," or something mystical, "the muses guide my fingers across the keyboard," but rarely practical. The most cited advice is to read everything about anything you can get your hands on, and while I applaud this technique, it is rather broad.
First, why research at all? One of the best pieces of advice I've seen on how to tell whether a book is going to be good or not--and unfortunately I forgot just where that came from--is to look at how the author treats horses. Do the horses the heroes ride get taken care of by anyone? Do they get tired? Do they carry knights in plate-mail endlessly across the horizon? Do they love their masters unconditionally and face off wolves for them and find them always in the nick of time no matter what? Do they, for love of god, talk?
The reason why you ought to care is because it shows that the writer was willing to invest a little time into our world as well as theirs. If they can't be bothered to do even the slightest bit of research, or don't have the needed prior knowledge and experience, why should I expect them to have put the same effort into their book? I can't be amazed at the differences in your story or world if I don't have a basis world to compare them with, and that world I know is the one I live in.
For instance, I read a book in which the horses talked, had true horse names, and unlimited energy. I got through this until I reached the part where all the horses followed their riders into a system of cave tunnels the humans had to duck to get through, and then on a spoken command leave at some point to go wait at the entrance for their return.
Of course, there are some exceptions to this. For one, your story might be about magical horses, ala Mercedes Lackey: in that case, you better have gotten the other details in your story right. Which leads to the "for two," which is that you don't need to use horses for this metric if you don't want to, especially if the space ballad you're reading (or writing) is equine free. Another really good one to look for is swords, particularly fencing related, and the use of scientific/mathematical terms. Misused jargon can be a clear warning sign. Don't even get me started on the portrayal of hacking in literature.
So if point one for "why research" is "so readers don't think you're a misinformed ingrate," than point two is "it's uncountably helpful." I'm sure you've heard the old adage that everything that's being written already has been (in fact, maybe because you've recently read one of my stories...) but while I feel this strictly isn't true, it's very possible that everything that's been written has already happened, at least in some form or another. Maybe Martian aliens haven't landed--but the Spanish ran into the unsuspecting South American civilizations in a similar way. Maybe fairies aren't real--but children have been kidnapped, switched at the hospital, and felt otherwise out of place. Finding out what happens in these cases can lend your story a degree of realism, practicality, and elegance that you just aren't going to get any other way.
How do you find out about these realistic things that are like what you're writing, anyway? I suppose that if I had to give a direction that would satisfy anyone, whether they are writing the next space thriller, werewolf romance, or mermaid court intrigue, is to read history. Psychology would be extraordinarily useful for anyone that has, you know, characters, and sociology if those characters ever interact with any other character, but books written on these subjects can be dreadfully dull. The good thing about history texts exists in what's called "original sources"--that is, first-person (mostly) accounts of what someone saw happen, or had happen to them. Philosophy is growing more important with the rise of post-modern stories, but again, often difficult to slog through. (Try to take classes in these instead, or alternatively post your favorite easy-to-read texts to this post.)
Right now I'm working on the Crown Jewel rewrite, so my floor is currently littered with books that I think will be useful (more on this later). Some of them will be much more useful than others, but you never know where you're going to get just the right detail. I was reading an article on Pompeii brothel theory, for instance, and the next piece was on the significance of the etchings on the back of Roman mirrors--of absolutely no relation to anything in the Jewel, although I think some of the characters may own mirrors--but since the significance ended up being prophecy, there were some elements I could clearly apply to my prophetic characters. I am trying to come up with a concrete government system for the Jewel, so I've been reading up on difference civilizations to see how they did it. I went and checked out almost every single book on fashion cira -500--1100 I could find. I am also reading some things I do not think will help me at all, such as Hildegard de Bingen's letters and Margery Kempe's autobiography... but you never know.
You could argue that the Jewel is a natural story for which to read history, because it's set in a pre-modern time. But the thing is that awesome people interactions happen in history. For instance, there was a fairly long time in which, because you could get divorced by claiming intercourse hadn't happened in a marriage, royal marriages had observers during their marriage night to make sure the deed was done. Can you think of all the stories that could be made off that one fact? The groom isn't quite sure what to do, so calls the adviser over...
Hmm, maybe I've got a story to work on now.