This follows on from my previous post on finding papers in the first place :-) Reading a paper is a bit of an art, in some ways, & going by my own experience something we're almost expected to pick up by osmosis. Well, I guess that's not completely true - at the Hons/Masters level, you're usually given a reading list to work through before each seminar, & then you sit around with the rest of the students in the class & your prof, & critique them. And that was where I really learned how to do it. But at school? Your teacher probably doesn't have time to spend on discussing the relative merits of a bunch of papers, or even on how to read them. And it's not just a matter of starting at the beginning & going on till you get to the end.
So - let's suppose that you've found a recent paper that looks like it might be relevant to your topic. (You've probably used a google search to do this; or even better, Google Scholar or the science-literature search engine scirus.) Where to start?
Well, don't go all out & read it from beginning to end. There are quicker ways to find out whether the paper is going to be useful to you (titles alone can sometimes be a bit misleading in that regard). But do start at the beginning: after the title, the abstract & the introduction are going to tell you a lot. The abstract, in particular, because it is meant to be a very concise summing-up that also highlights the significance of that particular bit of research. The introduction will help to put it all in context - why the researchers are looking at this particular point, why it's important, what is already known. Often I'll put the paper aside at that point, because it isn't really what I'm looking for.
Still looking good? At this point I usually jump to the discussion (others might go to the conclusions), which is where the authors are interpreting their results & explaining how they relate to that field, & expanding on the significance of what they've done.
And if the answer's still, yes, that's directly relevant to my question, then I'll go back & read it more thoroughly, paying attention to what was done & what the results were. I also look at the References section - you can find other relevant papers that way (& sometimes I can get quite happily sidetracked, following up on something I've spotted there). And as Keshav (n.d.) says - reading a good paper can also help you to learn how to write one.
Don't worry if you simply can't follow what the paper's talking about - I feel that way sometimes myself. This could be because I'm reading something in a field that's totally new to me; or I don't understand why they've used a particular method; or the paper's not particularly well written - or maybe it's the end of the day & my brain is tired! Nothing wrong with putting that paper aside & looking for another that's more accessible :-)
Oh yes, & not all papers are the same. Review papers are what the name suggests - they review a large body of work & pick out the key themes/concepts. These may be the most useful to you, because they do give you that broader coverage of the field, and give you some indication of what the consensus view is. Research papers report on a particular piece of work, & for a school student they're going to be harder to assess because you aren't necessarily familiar with the research area.
And to finish up, here are a couple of papers on how to read papers (!) - both freely available on the net:
S.Keshav (no date) How to read a paper. David R. Cheriton School of Computer Science, University of Waterloo, Waterloo ON, Canada; access date 22/06/2009
M.Pop (no date) How to read a scientific paper. Computer Science, Centre for Bioinformatics & Computational Biology, University of Maryland, USA; access date 22/06/2009