Writing an abstract that reviewers can assess
An abstract is the most-read part of any paper. Many readers will see only the abstract, and reviewers form an early impression from it before turning to the full text. Because it has to stand alone, it is judged by a different standard from the rest of the manuscript. Getting it right repays the effort twice over. This page explains what an abstract must contain and what you should confirm on the journal's official pages.
What an abstract is for
The abstract condenses a paper into a short block that lets a reader decide whether to read on. It is also what appears in databases and search results, so it shapes how the work is found. For that reason it must be understandable without the rest of the paper and must not rely on abbreviations that a general reader would not know.

It is a summary, not an advertisement. Its job is to report accurately, not to persuade.
What a strong abstract contains
A well-built abstract usually covers the background or problem, the objective, the methods, the main findings and the conclusion. Not every field uses every element, but the logic should be visible: this was the question, this is how it was studied, this is what was found, this is what it means.
Concrete detail is what makes an abstract credible. Statements that describe the work specifically are more useful than broad claims about importance.
What to leave out
An abstract should not contain citations to the literature, undefined abbreviations, or material that is not in the paper. It should not promise applications that the study did not test, and it should not hedge so heavily that no finding emerges. Tables, figures and footnotes are normally excluded.
Numbers should be reported with the same care as in the main text, including a measure of uncertainty where it applies. An abstract that reports a result without any indication of its reliability invites doubt.
Phrases that weaken an abstract
Certain habits reduce the impact of an otherwise good abstract. Vague openers such as a statement that a topic is important say little. Overclaiming, using words such as novel for routine work, harms credibility. Copying sentences directly from the introduction wastes the limited space.
Equally damaging is a mismatch between the abstract and the paper. If the abstract emphasises a finding that the results do not support, a reviewer will notice immediately, and trust in the whole manuscript suffers.
Structure and limits
Many journals require a structured abstract with labelled sections, while others prefer a single flowing paragraph. Word limits are usually enforced. The table below sets out the common features and what each one asks of you.
| Feature | What it provides | What to check |
|---|---|---|
| Background | Why the study was needed | Brief and specific |
| Objective | The question addressed | A single clear aim |
| Methods | How the work was done | Enough to judge design |
| Results | The main findings | Accurate, with uncertainty |
| Conclusion | What the findings mean | Within the evidence |
Writing it in practice
The simplest method is to write the abstract last and work backwards from the paper. Draft one sentence for each key element, then tighten until the limit is met. Read it aloud to catch awkward phrasing, and check that every sentence would make sense to a reader who has not seen the paper.
Ask whether a specialist in another subfield could understand it. If not, remove or define the jargon. If a sentence merely repeats another, cut it.
Checking it against the paper
Before submission, compare the abstract with the manuscript line by line. Every claim in the abstract should be traceable to a result, and every important result should appear. This check catches the mismatches that reviewers most often flag.
- Does the abstract state the same objective as the introduction?
- Are the reported findings consistent with the results section?
- Does the conclusion avoid claims the data cannot support?
- Are abbreviations defined or avoided?
- Does it fit the journal's word limit and required format?
Keywords and title together
Most journals ask for keywords alongside the abstract. These help indexing, so they should reflect the paper's actual content rather than a wish list of popular terms. Together, the title, abstract and keywords form the surface that readers and search tools use to find the work.
Because format, word limits and required elements vary, check the journal's instructions before you finalise your abstract. The current requirements are the ones set by the publisher, so confirm them on the journal's official pages rather than relying on a general guide.
An abstract must stand alone, report the study accurately and fit the journal's format. Cover the problem, aim, methods, findings and conclusion, avoid overclaiming, and make sure it matches the paper exactly. Word limits and structure are set by the publisher, so confirm the current requirements on the journal's official pages.
Read next
How to structure a research manuscript
A conventional section-by-section layout that reviewers can follow.
References and citation style: getting them right
Why accurate referencing matters and how to apply a journal’s style consistently.
Figures, tables and units in a scientific paper
Presenting data clearly, labelling correctly and avoiding duplication.