How to Write an Abstract for a Research Paper
A complete guide to writing a strong, journal-ready abstract, with structure breakdown, field examples, and the most common mistakes to avoid.
Guide
An abstract is your paper's first impression. Editors use it for desk rejection decisions. Reviewers use it to frame their evaluation. Readers use it to decide whether to read your paper. Writing a strong abstract is one of the highest-value skills in academic publishing, and one of the least explicitly taught. This guide breaks down exactly what belongs in an abstract, how to structure it, and how to avoid the mistakes that weaken most first drafts.
The five elements of a complete abstract
Every research abstract must include five elements, regardless of discipline:
Background/context: Why does this research matter? What problem does it address?
Objective: What specific question does the paper answer or what did you set out to do?
Methods: How did you study this? Study design, data source, key techniques.
Results: What did you find? Include specific numbers where possible.
Conclusions/implications: What do the findings mean? What's the take-home message?
Word count and format requirements
Most journals require 150–300 words for standard abstracts. Structured abstracts (with explicit subheadings) are required by many medical, clinical, and some social science journals, and can run up to 350–400 words. Always check your target journal's specific requirements before writing, the word limit is in the 'Instructions for Authors' page.
Structured vs unstructured abstracts
Structured abstracts use explicit subheadings (Background, Methods, Results, Conclusion) as required by journals like JAMA, Lancet, and many clinical journals. Unstructured abstracts are continuous paragraphs, typical in most other fields. The content requirements are the same, only the formatting differs.
Common abstract mistakes
The most common abstract errors that lead to rejection or reviewer frustration:
Missing the main result, abstract describes the study but omits what was found
Vague claims, 'significant differences were observed' without numbers
Too long, exceeding word limits is a formatting violation
Citations in the abstract, most journals prohibit them
Introducing new content not in the paper, abstract must match the full text
Starting with 'In this paper...', weak opening that wastes words
Field-specific examples
Different disciplines have different abstract conventions. Biomedical papers typically use structured abstracts with exact patient numbers, statistical results (p-values, confidence intervals), and outcome measures. Computer science abstracts often lead with the problem statement and contribution, then describe the technical approach. Social science abstracts often include the theoretical framework and broader implications alongside the empirical findings.
When to write your abstract
Write your abstract last. Once your paper is complete, you can accurately summarize what you found, how you found it, and what it means. Many researchers write a draft abstract first to guide their writing, then revise it substantially at the end to reflect the actual completed paper.
Ready to check your manuscript?
Upload your paper and get a submission readiness report in under 2 minutes.
Check my manuscript, it's freeNo account · PDF, Word, LaTeX · Results in <2 min
Frequently asked questions
Standard abstracts: 150–300 words. Structured abstracts: up to 350–400 words. Conference abstracts: typically 100–250 words. Always check your target journal's specific word limit.
This depends on journal style. Many STEM journals accept 'we' (e.g., 'We found that...'). Others prefer passive constructions. Check examples in your target journal.
Background and context: present tense. Methods and results: past tense. Conclusions and implications: present tense. This pattern is standard but varies by discipline.
Not typically, abstracts focus on what was done and found. Limitations belong in the Discussion section. Some journals include a brief limitation statement in the abstract conclusion, but this is not standard.