01Most of what you open should be put down
Students read research articles front to back, once, slowly, feeling guilty at every paragraph that will not stay in their heads. Researchers do not read that way at all. They read in passes, and each pass exists to answer one question: is this worth the next pass.
Abandoning an article is not failure. It is the correct outcome for most of them, because a search returns everything on a topic while your paper needs only the studies that actually bear on your question. The cost of reading a poor match all the way through is not the hour it took; it is the good article you never reached because the hour was gone.
So the order below is designed to let you quit early and know why. Two minutes, then ten, then the whole thing, with a decision at the end of each.
02First pass: two minutes, and mostly a no
Read the title, the abstract, and then the final paragraph or two of the discussion. That last piece is the part almost everybody skips, and it carries more information per line than anything else in the paper, because it is where the authors say plainly what they believe they have shown and what they have not.
You are answering one question: does this study concern the same population, the same intervention or exposure, and the same kind of outcome as the question I am writing about. Not the same topic. The same question. An excellent study of a different population is a poor source for your paper and a fine one for somebody else's.
If the answer is no, stop, and write one line saying why before you close it. That line is what prevents you opening the same article again in three weeks having forgotten you rejected it, which is an hour every student loses at least twice. It is also the moment to record where the copy came from, since a source you cannot find again is a source you cannot cite, and there is more on where open copies actually sit.
03Second pass: the methods, before anything persuasive
If it survives, go to the methods section next, and go there before you read the introduction. The introduction is written to persuade you that the study matters; the methods tell you whether it can support what it claims, and reading them first keeps you from being convinced by an argument you have not yet checked.
Look for four things: who took part and how they were chosen, what was actually done to them or measured about them, over what period, and what was compared with what. Then go to the tables and read them before the results text, since the text describes the numbers selectively and the tables do not.
Here you make the second decision. Does this study's evidence do the job your sentence needs. A great deal of the time you now have what you came for, which is one finding, one figure and one citation, and there is no reason to read the rest of it at all.
04Third pass, and writing it down once
The full reading is for the small number of articles that sit at the centre of your argument: the ones you will discuss rather than merely cite. Read those properly, introduction included, and read the limitations section slowly, because it will tell you what the authors already know is weak and will save you defending something they never claimed.
Whatever pass you stop at, write the note before you close the window. Rereading because the note was too thin is the single largest waste of time in a literature review, and a note written while the article is open takes two minutes against the twenty it costs to reconstruct. Put the note in the same entry that holds the citation itself, which keeps the two together and heads off the problem described in reference managers and citation drift. Where the reading itself is what a term cannot hold, that stage can be carried alongside a longer research project.
Five lines per article is enough, and the same five every time:
- What the study set out to answer, phrased the way you would phrase it rather than the way they did.
- Who was studied, how many, and over what period, in one line.
- The single finding you would actually use, with the number and the page or table it sits in.
- The weakness that matters most, usually taken straight from the limitations section.
- Where it goes in your own paper, named by section, or the reason you rejected it.
FAQQuestions to control
What order should I read the sections of a paper in?
Abstract and the closing paragraphs of the discussion first, to decide whether to continue. Then methods, then tables, then the results text. The introduction comes late and often not at all, since it exists to persuade rather than to inform. Read the limitations section carefully for any article you intend to build an argument on.
When is it right to put an article down?
When you can say what it would contribute to your paper, or when you can say why it will not. Both are complete answers. Most articles resolve at the two minute pass because the population, the intervention or the outcome does not match your question. Write one line recording the decision either way, or you will open the same article again later.
Why read the methods before the introduction?
Because the introduction is an argument for the study's importance and the methods are the evidence for whether it can deliver anything. Reading the argument first makes you a more forgiving reader of the design. Taking the methods cold, before anyone has told you why it matters, is how you notice that a sample was small or a comparison group absent.
How many articles should I actually read in full?
Far fewer than you cite. A literature review might rest on a handful of studies you have read completely, discussed in detail, with a wider set cited from a single finding each. That is normal practice rather than a shortcut. The judgement being assessed is which studies deserved the close reading, not how many hours you spent reading indiscriminately.
