Note-Taking and Bibliographic Records
Capture the full reference the moment you find the work. Every hour spent reconstructing citations at the end is an hour taken from the argument.
The rule that saves the most time
The word full is doing work too. A partial capture — author and title, no volume, no page range — is a reference you will have to return to the database for. Capture every element the pattern requires, in one pass. Harvard reference anatomy sets out exactly which elements those are for journal articles and books.
What the source tells you to record
The Week 4 record-keeping sequence
Full bibliographic details, immediately
Every element of the reference, in the Harvard pattern, at the moment you find the work. The source names reference-management software by category — "Endnote or similar programs" — without recommending a specific one.
Skim for relevance and quality
"Read the literature quickly (skim read) and make some short notes as to its relevance and quality." A short note, not a summary. This pass decides whether the work earns a careful read.
Read carefully, annotating as you go
"Carefully read the literature, annotating sources and noting the value of the reading to your topic." The value to your topic is the part most people leave out and most need later.
Write your thoughts down daily
"As you go along, write down your thoughts. Write daily if possible, even if it only a paragraph or two. Do this first thing each day if you can."
Date every draft and reference
"Note and date all your drafts and references as you go along." Without dates you cannot tell which version of an argument is current.
Back up in a separate location
"Make sure you keep back-up copies of your writings and store copies in a separate location e.g. a portable hard drive." The source's emphasis is on separate — a second copy in the same place is not a backup.
Why writing daily is a research instruction, not a productivity tip
The daily-writing rule looks like generic study advice and is easy to skip. It is doing something more specific. Reading produces an understanding that feels stable and is not; the act of writing a paragraph is the only reliable test of whether you actually followed the argument you just read.
It also front-loads the review. The source's own account of the theoretical framework describes it as something you construct from the literature — see The conceptual and theoretical framework. A file of daily paragraphs written while the reading was fresh is raw material for that construction. A folder of unannotated PDFs is not.
What daily writing produces
- A record of what you understood at the time, in your own words
- Early detection of arguments you cannot actually reconstruct
- Draft sentences that survive into the review with light editing
- A dated trail showing how your position changed and why
What a folder of PDFs produces
- A false sense that the reading is done
- Re-reading, because nothing was extracted the first time
- Citation reconstruction under deadline
- No evidence of how your thinking developed
The instruction worth arguing with
A record structure that survives to the write-up
The source specifies no format. The structure below is a synthesis for this page. Its only design goal is that every work is recorded on the same axes, because comparability is what turns notes into a review rather than a list.
ONE RECORD PER WORK
| Field | Captured when | Why it has to be there |
|---|---|---|
| Full reference in the Harvard pattern | On discovery | The source's explicit instruction; the only field that is cheap now and expensive later |
| Where you found it, and the search that found it | On discovery | Lets you re-run or defend the search; lets you find the sibling papers |
| Relevance and quality, one line | Skim pass | Decides whether a careful read is warranted |
| Central definitions, verbatim, with page numbers | Careful read | Paraphrased definitions drift and hide genuine disagreements between authors |
| Methodology, sample, and how the sample was obtained | Careful read | The only way to weigh two studies against each other |
| Stated limitations | Careful read | Required by the source's own annotation anatomy, and omitted from its reading questions |
| Value to your topic, in your words | Careful read | The source asks for this explicitly and it is the field most often skipped |
| Date of the note | Always | Makes the trail auditable and shows which version of your thinking is current |
Rows one, three and seven are the source's instructions. The rest is a synthesis for this page.
What good records let you produce
Records built this way feed three separate deliverables without further reading. They supply the annotations if you are asked for an annotated bibliography — see Writing an annotation. They supply the grouped positions that become the body of a review — see Writing the literature review. And the methodology and sample fields feed your own design decisions rather than your review at all.
What to carry forward
- Capture the full reference at the moment of discovery. This is the single highest-return habit in the whole process.
- Two passes: skim for relevance and quality, then read and annotate for value to your topic.
- Write a paragraph daily. It is a comprehension test, not a productivity tip, and it produces reusable draft.
- Date everything and keep a copy somewhere physically separate.
- Treat "articles rather than books" as advice about currency, not a quality rule — the same source's definition of literature includes books.
Frequently asked questions
Which reference manager should I use?
The supplied source names one program by way of example and adds "or similar programs" — it does not recommend a specific tool or compare them. Any manager that stores full Harvard elements and exports cleanly will satisfy the instruction; the discipline of capturing on discovery matters more than the software.
How detailed should notes be on the first pass?
Short. The source is explicit that the skim pass produces "some short notes as to its relevance and quality" only. Full characterisation belongs to the careful second read, and doing it on everything wastes time on sources you will discard.
Is 'select academic articles rather than books' a real rule?
It is in the source, stated without qualification, but it sits against the same week's own definition of literature, which explicitly includes books. Read it as advice about currency and specificity for a time-limited project, not a quality ranking.
Why does the source insist on dating drafts?
So you can tell which version of an argument is current, and so the development of your thinking is auditable. It matters most when a position changes late and you need to know which sections still reflect the old one.
What should I record that the source does not ask for?
Stated limitations, and the search that found the work. The reading questions omit limitations even though the annotation anatomy requires them, and recording the search lets you defend or re-run it later.
References and source attribution
- Supplied teaching source, Week 4: Reviewing the Literature — study notes and slide deck.
- Supplied teaching source, Week 3: Developing a Research Topic — the literature-checking routine.
- O'Leary, Z. 2017, The Essential Guide to Doing Your Research Project, 3rd ed., Sage Publications, London.
- Bryman, A. 2016, Social Research Methods, 5th ed., Oxford University Press, Oxford.
Suggested questions for Ask KEVOS
- Design me a literature note template based on the fields on this page.
- What are the full Harvard elements I need to capture for a journal article?
- How do I keep notes that feed both an annotated bibliography and a literature review?
- What is the risk of paraphrasing a definition in my notes?
- How should I organise notes when I have more than fifty sources?