01Where the summary stops and the annotation begins
An annotation makes three moves in order. Say what the source did, which is summary. Judge how sound it is, which is appraisal. Then say what it does for the paper you are going to write, which is the move that makes the exercise worth setting at all. Most entries perform the first at length, wave at the second, and leave the third out.
That third move is also why this lands before the essay rather than after it. The assignment is a rehearsal for an argument: after twenty entries you should be able to see which sources agree, which quarrel with each other, and where nobody has looked yet. If your entries could be shuffled into any order without a single word changing, that thinking has not happened, and a marker can tell from a page away.
02The moves, in the words markers use
Write each entry as one paragraph that walks through these in sequence. The proportions matter as much as the content: if summary is taking more than half your space, the entry is out of shape.
- What it did. Design, participants or texts, setting, and the finding, in your own words and in about two sentences. Not the published abstract with the vocabulary swapped
- How sound it is. Not whether you enjoyed it. How the evidence was gathered, how far it reaches beyond its own sample, when it appeared, and what the authors themselves list as limitations
- What it does for you. The claim it supports, the hole it reveals, an instrument you may borrow, or the counter-argument you will have to answer somewhere in chapter two
- Where it sits beside its neighbours. One clause tying it to another source in the list is what turns a pile of entries into a body of literature you can write from
03The mechanical marks nobody should be losing
These are the criteria that get scored fastest and grumbled about most, and every one of them is settled before you write a sentence about content.
- References formatted in the style your brief names and ordered alphabetically unless it says otherwise
- Entries of roughly even length. Wildly uneven ones read as unevenly read
- Primary research where the brief asks for it, since a review article is not a study and a textbook chapter is neither
- Publication dates inside whatever window is set, with older work kept only where you say in a clause why it still stands
- Past tense for what the authors did, present tense for what the work argues, held consistently down the whole list
04Reading twenty sources without reading twenty papers
The panic about workload is understandable and mostly unnecessary, because you are not reading these end to end. Take the abstract to decide whether a paper stays. Take the method section to judge it. Take the discussion for what the authors believe it means. Only open the results if your appraisal turns on them. Introductions are the most quoted and least useful pages in a paper for this particular job.
Two habits carry the rest. Sort as you go into three piles, one for work that supports your argument, one for work that complicates it, and one for methods you might borrow, because those piles become paragraphs later. And write each entry the day you read the source, since an annotation attempted a fortnight afterwards collapses back into summary: by then the only thing you still remember is roughly what the paper was about. Where the reading list is longer than the term, a paper and its sources can be built alongside you, and the doctoral version of this same problem is handled as chapter work.
FAQQuestions to control
How long should each annotation be?
Whatever your brief says, and if it says nothing, one tight paragraph of roughly one hundred to two hundred words is the usual expectation. Consistency matters more than the number itself. Twenty entries of similar length signal that every source was given the same attention, whereas three long ones followed by seventeen thin ones tells a marker precisely where you ran out of evening.
Can I write the annotation from the abstract?
No, and it shows almost immediately. An abstract is the authors' own summary, which means it skips the two things you are being marked on: your appraisal of how the evidence was produced, and the relevance of the work to your argument. It also invites close paraphrase of somebody else's sentences. Open the method and discussion sections; that is where your entry actually comes from.
Does an annotated bibliography need an introduction?
Many briefs ask for a short one, and it is worth writing even when they do not. Two or three sentences naming your research question and the criteria you used to select sources will do: databases searched, date range, types of study kept and rejected. It frames the list as deliberate rather than assembled, and it becomes the search description your methodology section needs later.
Two of my sources say almost the same thing. Keep both?
Keep both only if you can say why they are not redundant, which is often possible: different populations, different settings, or one replicating the other in a way that strengthens the claim. Otherwise keep the stronger study and note in the weaker entry that its finding is confirmed elsewhere. Making that judgment visible is worth marks, because comparison is the skill being tested.
