How to Write an Abstract for a Research Paper

What the abstract has to do

Almost everyone who encounters your paper will read only the abstract. It is what appears in database search results, what a reviewer skims before agreeing to review, and what a busy reader uses to decide whether the full text is worth twenty minutes. So the abstract has one job: make the paper's contribution legible to someone who has not read it.

That means it must stand alone. No "as discussed below," no undefined abbreviations, no figure references, no dangling "this study addresses an important gap" without saying which gap.

The five moves

Nearly every good abstract in any discipline makes the same five moves, in the same order. Write one or two sentences for each:

1. Context — the one thing a non-specialist needs to know to care.
2. Gap or problem — what is unknown, contested, or broken.
3. What you did — design, data, sample, method, in concrete terms.
4. What you found — the actual result, with the direction and size of the effect.
5. What it means — the implication, stated without overclaiming.

Nature's formatting guide describes a version of this same shape for its summary paragraph: two or three sentences of basic-level introduction to the field, a brief account of the background and rationale, a statement of the main conclusions introduced by the phrase "Here we show" or its equivalent, and two or three sentences putting the findings into general context. It is a useful template even if you will never submit to Nature.

Before and after

Here is a first draft that makes all five moves badly:

Sleep is important for students. This study investigates the relationship between sleep and academic performance. A survey was conducted and the data were analysed. The results are discussed and implications for practice are considered.

Nothing here is false — and nothing is informative. There is no sample, no measure, no finding. Compare:

Short sleep duration is common among undergraduates, but its relationship to academic performance is often confounded by workload. We surveyed 412 first-year students at a single university across one semester, pairing self-reported sleep diaries with end-of-term grades and controlling for weekly paid work hours. Students averaging under six hours per night scored 0.31 grade points lower than those averaging seven to eight hours, and the association held after adjusting for work hours. The effect is modest but persistent, suggesting sleep is worth including in models of academic performance alongside workload.

Same length. The difference is entirely in specificity: a number, a method, a direction, a caveat. If a sentence in your abstract would still be true of a completely different paper, it is wasted.

Word limits and formats: check before you write

Abstract requirements are set by the venue, not by convention, and they vary more than most writers expect:

• APA Style notes that journal abstracts are typically limited to no more than 250 words. In APA papers the abstract sits on its own page with the labelAbstract centered and bold, the first line flush left rather than indented, followed by three to five keywords on an indented line beginning with the italicised label Keywords:.
• Nature asks for a fully referenced summary paragraph, ideally no more than 200 words, aimed at readers outside the discipline and avoiding abbreviations and acronyms unless essential.
• Medical and health journals typically require a structured abstract with labeled sections, and often allow more words to accommodate them.

Structured vs. unstructured

A structured abstract splits the same five moves under explicit headings — Background, Methods, Results, Conclusions — while an unstructured abstract runs them together as one paragraph. Don't guess which your journal wants; the author instructions will say, and submitting the wrong format is an easy desk-rejection.

If your paper follows a reporting guideline

Some study types have an abstract checklist of their own. PRISMA 2020 for Abstracts specifies 12 items a systematic review abstract should report — including the objective, eligibility criteria, information sources, risk-of-bias assessment, number of included studies, synthesis methods, limitations, registration, and funding. CONSORT for Abstracts plays the same role for randomised trials, requiring a structured summary of trial design, methods, results, and conclusions. If your paper falls under one of these, work through the checklist line by line before you submit — reviewers do.

Five mistakes reviewers notice first

• No result. "Findings are discussed" is not a finding. Give the number or the direction.
• Undefined abbreviations. The abstract is indexed separately from the paper, so a term you defined in the introduction is undefined here.
• Overclaiming. A correlational design cannot "show that X causes Y." Match the verb to the evidence — associated with, predicted, was consistent with.
• Copy-paste from the introduction. An abstract that is three sentences of background and one of method is a background paragraph wearing a hat.
• Drift from the paper. Numbers get revised during peer review and abstracts often don't. Re-check every figure in the abstract against the final results section.

A last pass before you submit

Read the abstract aloud with the rest of the paper covered. If you can't tell what was done, what was found, and why it matters, neither can a reader. Then check the word count against the author instructions — abstracts are one of the few places where a submission system will simply refuse to accept an overlong text.

Frequently asked questions

Keep reading

• How to Write a Literature Review Without Fabricating Sources
• How to Respond to Reviewer Comments
• Why ChatGPT Makes Up Citations
• Free APA Citation Generator