The Five Pillars
One of the most important essays in Wikipedia is WP:FIVEPILLARS which is designed to eloquently sum up what we're here for.
Once you get your head around these five pillars, you will be a Wikipedian and a good one at that. All 5 are covered in my adoption school, though at different lengths. Be aware that I don't know everything and I would doubt anyone who said they did.
How articles should be written
The articles in Wikipedia are designed to represent the sum of human knowledge. Each article should be written from a neutral point of view – personal opinions such as right and wrong should never appear, nor should an editor's experience. Neutrality also means giving due weight to the different points of view. If the broad scientific community has one set of opinions, then the minority opinion should not be shown. An example is in medicine: if there was an article on, say treatment of a broken leg, a neutral article would not include anything on homeopathy.
To ensure that the information in an article is correct, Wikipedia has adopted a policy of verifiability. Anything written in Wikipedia should be available to confirm by looking at the associated reliable source. Wikipedia should not include anything not verifiable by seeing it is published elsewhere; in other words, it should not contain anything original.
So what is a source? Wikipedia uses the word source for three interchangeable ideas – a piece of work, the work's creator or the work's publisher. In general, you would expect a reliable source to be published materials with a reliable publication process, authors who are regarded as authoritative in relation to the subject, or both. This doesn't mean that a source that is reliable on one topic is reliable on every topic, it must be regarded as authoritative in that topic. So whilst "Airfix monthly" may be a good source on the first model aeroplane, I would not expect it to be authoritative on their full-size equivalent.
A source that is self-published is, in general, considered unreliable, unless it is published by a recognized expert in the field. This is a very rare exception, so self publishing is generally considered a no-no. This means that anything in a forum or a blog and even most websites are considered unreliable by default. One interesting sidepoint is on self-published sources talking about themselves. Obviously, a source talking about itself is going to be authoritative, but be careful that the source is not too self-serving; the article really should not be totally based on a direct source like that.
Mainstream news sources are generally considered reliable... but any single article should be assessed on a case-by-case basis. Some news organizations have been known to check their information on Wikipedia – so be careful not to get into a cyclic sourcing issue!
There's a lot more about what makes a source reliable here.
Any questions or would you like to try the test?