|
Getting your Trinity Audio player ready...
|
Key Takeaways
- An effective abstract answers four questions in miniature: why the study was done, how it was done, what was found, and what it means.
- Word count limits, structure rules, and formatting conventions vary by publication type, so always check the specific journal, conference, or institutional guidelines before submitting.
- Cutting words without cutting meaning usually means removing hedges, background padding, and redundant phrases, not removing data or conclusions.
Contents
- Glossary of Key Terms
- What Is an Abstract and Why Does It Matter?
- What Are the Main Types of Abstracts?
- How Do You Write an Abstract Step by Step?
- Annotated Abstract Examples from Different Disciplines
- How Can You Reduce Your Abstract Word Count?
- Dissertation, Journal, and Conference Abstracts: What Is the Difference?
- How Do You Tailor Your Abstract to Your Target Journal?
- Abstracts for Narrative Reviews, Systematic Reviews, and Scoping Reviews
- How Do You Write an Abstract for a Study with Null Results?
- Abstracts for Stage 1 and Stage 2 Registered Reports
- Frequently Asked Questions
Glossary of Key Terms
The terms below appear throughout this guide. Refer back to this list whenever a term is unfamiliar.
- Abstract: A concise, standalone summary of a paper’s purpose, methods, results, and conclusions, typically 150 to 300 words long.
- Structured abstract: An abstract divided into labeled subsections, such as Background, Methods, Results, and Conclusion.
- Unstructured abstract: A single continuous paragraph that summarizes a paper without labeled subsections.
- IMRaD: An acronym for Introduction, Methods, Results, and Discussion; the standard structure that many empirical papers and their abstracts follow.
- Keywords: A short list of terms placed after an abstract to help indexing databases and search engines categorize the paper.
- Narrative review: A qualitative synthesis of existing literature on a topic, written without a predefined, replicable search protocol.
- Systematic review: A review that uses a predefined, documented search and selection protocol to answer a specific, focused research question.
- Scoping review: A review that maps the breadth of evidence on a topic, often used to clarify concepts, terminology, or research gaps.
- Registered Report: A publication format in which the study protocol is peer reviewed and receives in principle acceptance before data collection or analysis begins.
- Null result: A finding in which no statistically significant difference or relationship is detected between the variables studied.
- PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses; a reporting checklist commonly used for systematic review abstracts and papers.
- Preregistration: The practice of publicly documenting a study’s hypotheses and methods before data collection begins.
What Is an Abstract and Why Does It Matter?
An abstract is a self-contained, 150-to-300-word summary that lets a reader decide, within seconds, whether the full paper is relevant to their work. It is usually the most read part of any paper.
Because most databases, search engines, and conference programs index only the title and abstract, the abstract effectively markets the paper. A well-written abstract increases the odds that a paper is found, read, cited, and, for conference submissions, accepted.
Beyond visibility, the abstract performs several functions at once:
- It summarizes the research question, methods, results, and conclusions in miniature.
- It helps editors and reviewers screen submissions quickly and consistently.
- It gives indexing databases the keywords and framing needed for accurate categorization.
- It sets reader expectations for the level of detail and the scope of the full paper.
What Are the Main Types of Abstracts?
The main types are descriptive, informative, structured, unstructured, critical, and graphical abstracts, and each serves a different publication context.
- Descriptive abstract: outlines the topic and scope of the paper without reporting specific results; common in humanities and conference programs.
- Informative abstract: summarizes the actual findings and conclusions, not just the topic; the standard for most empirical journal articles.
- Structured abstract: uses labeled headings such as Background, Methods, Results, and Conclusion; common in clinical, biomedical, and many social science journals.
- Unstructured abstract: presents the same content as a single flowing paragraph; common in humanities, some engineering venues, and many conferences.
- Critical abstract: adds a brief evaluative comment on the validity or reliability of the work; rare, used mainly in some review or commentary contexts.
- Graphical or visual abstract: a single image or diagram that summarizes the study, submitted alongside, not instead of, the written abstract.
How Do You Write an Abstract Step by Step?
Write the abstract last, after the full paper is complete, and build it from five components: background, objective, methods, results, and conclusion.
Background and Context
State the problem or gap in one to two sentences. Avoid a lengthy literature review; a single sentence establishing why the topic matters is usually enough.
- Useful sentence starters: “Despite growing interest in…”, “Little is known about…”, “Prior work has not addressed…”
Objective or Purpose
State exactly what the study set out to do, using a direct verb such as examine, test, compare, or evaluate.
- Useful sentence starters: “This study examined…”, “We tested whether…”, “The aim of this paper was to…”
Methods
Report the design, sample or data source, and key procedure in enough detail that a reader can judge the evidence’s strength, without listing every instrument or statistical test.
- Include: study design, sample size or corpus size, setting, data type, key variables, and the primary analytic approach.
- Omit: minor covariates, full instrument names, and secondary analyses.
Results
Report the primary finding with a specific number, direction, or effect, not a vague statement such as “results are discussed.”
- Weak: “The relationship between variables was significant.”
- Strong: “Sleep duration was positively associated with test performance (r = 0.34, p < 0.01).”
Conclusion
State what the findings mean and, where relevant, one practical or theoretical implication. Avoid introducing new data or citations in this final sentence.
Annotated Abstract Examples from Different Disciplines
Excerpts below are illustrative composites, shortened with ellipses, and annotated to show how each discipline applies the same core structure differently.
Discipline | Example Excerpt | Annotation |
Biomedical or clinical | “Background: Adherence to statin therapy remains suboptimal… Methods: In a randomized, controlled trial of 412 adults… Results: The intervention group showed a 14% higher adherence rate (95% CI, 8 to 20)… Conclusion: A brief pharmacist-led counseling protocol improved adherence.” | Structured, IMRaD format; reports exact sample size and confidence interval rather than only a p-value; conclusion is a direct claim tied to the numeric result. |
Psychology or social science | “This study examined whether perceived social support buffers the effect of workplace stress on burnout… Across two samples (N = 210, N = 185), support moderated the stress-burnout relationship… Findings suggest targeted support interventions may reduce burnout risk.” | Unstructured paragraph but still follows objective, method, result, conclusion order; hedged language (“suggest”) signals correlational, not causal, design. |
Computer science or engineering | “We propose a lightweight scheduling algorithm for edge computing environments… Evaluated on three benchmark datasets, our method reduces latency by 22% compared with the baseline while using 30% less memory… Code is available upon request.” | Leads with the contribution, not the background; quantifies improvement against a named baseline; typically unstructured and written in present tense. |
Humanities | “This essay reconsiders the role of epistolary form in eighteenth-century fiction… Drawing on close readings of three novels, it argues that the letter format destabilizes narrative authority… The analysis contributes to ongoing debates about voice and reliability.” | Descriptive, argument-driven abstract; no numeric results because the paper is interpretive; the verb “argues” signals the paper’s thesis rather than a finding. |
Environmental science | “Coastal wetlands provide critical carbon storage, yet degradation rates remain poorly quantified… Using satellite imagery from 2005 to 2022, we estimate a 9% loss in wetland extent… These losses translate to an estimated 1.2 million tons of stored carbon released.” | Uses a specific date range (a hyphenated numeric range) and a quantified impact statement; background is limited to a single problem-framing sentence. |
How Can You Reduce Your Abstract Word Count?
Cut word count by removing throat-clearing phrases, hedging words, and background padding, and by converting long phrases into single words, all without deleting data or conclusions.
- Cut throat-clearing openers: delete phrases such as “It is well known that” or “This paper attempts to”; start with the actual content.
- Trim hedging words: words like “fairly,” “quite,” “somewhat,” and “in order to” rarely add meaning and can usually be deleted.
- Replace phrases with single words: “due to the fact that” becomes “because”; “a large number of” becomes “many.”
- Limit background to one sentence: reviewers already know the general field; state only the specific gap your study addresses.
- Remove citations: most style guides discourage in-text citations within abstracts; move any essential attribution to the introduction.
- Avoid repeating the title: do not restate the paper’s title in the opening sentence of the abstract.
- Use numerals for numbers: write “3 groups,” not “three groups,” to save characters where the target style guide allows it.
- Combine short sentences: merge two related short sentences into one compound sentence using “and” or a semicolon.
- Delete redundant qualifiers: phrases such as “completely eliminate” or “absolutely essential” can drop the first word without losing meaning.
Dissertation, Journal, and Conference Abstracts: What Is the Difference?
The three differ mainly in length, audience, and purpose: dissertation abstracts are longest and most comprehensive, journal abstracts are the most tightly word-limited and indexed, and conference abstracts double as a submission that is itself peer reviewed for acceptance.
Feature | Dissertation Abstract | Journal Abstract | Conference Abstract |
Typical length | 350 words, sometimes up to 1 page | 150 to 300 words, per journal limit | 150 to 300 words, per call for papers |
Primary audience | Committee members and future researchers in the field | Journal readers and database indexers | Conference reviewers and attendees |
Main purpose | Summarize the entire dissertation’s scope and contribution | Summarize one study for search and screening | Persuade reviewers to accept the presentation or poster |
Common structure | Unstructured, comprehensive overview of all chapters | Structured or unstructured, per journal style | Often structured, with an explicit results requirement |
Review process | Approved with the dissertation by the committee | Reviewed alongside the full manuscript | Often reviewed independently, before the full paper exists |
How Do You Tailor Your Abstract to Your Target Journal?
Tailor an abstract by matching its structure and terminology to whether the journal is a narrow specialty title or a broad interdisciplinary one; specialty journals reward technical precision, while interdisciplinary journals reward accessible framing.
Specialty Journals
- Use field-specific terminology and abbreviations without defining every term; the readership already knows them.
- Follow the journal’s required structured format exactly, including its specific subheading labels and order.
- Lead with the most technically precise version of the result, including exact statistics or measurements.
Interdisciplinary Journals
- Define specialized terms on first use, or replace them with plain-language equivalents where possible.
- Open with the broader problem or question before narrowing to the specific study, since readers may not share your subfield’s assumed context.
- State the real-world or cross-field significance of the finding explicitly, rather than assuming readers will infer it.
Before submitting, read three to five recently published abstracts from the target journal and match their sentence length, structure, and level of technical density.
Abstracts for Narrative Reviews, Systematic Reviews, and Scoping Reviews
The three review types require different abstract content because they differ in how systematically the literature was searched and what question each answers.
Feature | Narrative Review | Systematic Review | Scoping Review |
Search protocol reported | Not required; search strategy is usually not described | Required in detail, including databases and dates | Required, though often less exhaustive than systematic |
Typical question | Broad overview of a topic or field | Focused, answerable question, often framed as PICO | Broad mapping question about scope, extent, or gaps |
Reporting checklist | None standard | PRISMA for systematic reviews | PRISMA extension for scoping reviews, PRISMA-ScR |
Abstract typically includes | Scope, main themes, and the author’s synthesis or argument | Number of studies screened and included, pooled or synthesized results | Number of sources included and a description of evidence characteristics, not pooled results |
Narrative Review Abstracts
State the topic, the general scope of literature covered, and the paper’s central argument or organizing theme. Because there is no fixed protocol, avoid implying a systematic search was performed.
Systematic Review Abstracts
Report the number of studies screened and included, the databases searched, the date range, and a synthesized result, ideally with an effect size or clear directional finding across studies.
Scoping Review Abstracts
Report the number of sources included and describe the range or characteristics of the evidence, such as study designs or populations covered, rather than a pooled outcome, since scoping reviews map a field instead of testing a hypothesis.
How Do You Write an Abstract for a Study with Null Results?
Report null results with the same specificity as significant ones: state the effect size and confidence interval, avoid words such as “failed,” and frame the finding as informative rather than as a shortcoming.
- Report precision, not just significance: give the confidence interval or effect size so readers can judge whether the study was adequately powered to detect an effect, not only whether p exceeded 0.05.
- Avoid apologetic language: phrases like “unfortunately” or “failed to find” frame a null result as a defect; “no significant association was observed” is neutral and accurate.
- State the practical implication: explain what the absence of an effect means for theory or practice, such as ruling out a proposed mechanism.
- Mention preregistration if applicable: noting that hypotheses were preregistered strengthens the credibility of a null finding by ruling out selective reporting.
- Note power and limitations briefly: one clause on sample size or power helps readers distinguish “no effect” from “underpowered to detect an effect.”
Example: “Contrary to the preregistered hypothesis, cognitive training showed no significant effect on working memory scores (d = 0.04, 95% CI, negative 0.12 to 0.20). The finding suggests the intervention’s benefits, if any, do not generalize beyond the training task itself.”
Abstracts for Stage 1 and Stage 2 Registered Reports
A Registered Report abstract changes with each stage: the Stage 1 abstract summarizes a planned study with no results yet, while the Stage 2 abstract adds the actual results after data collection is complete.
Stage 1 Registered Report Abstract
Written before data collection, so it cannot include results. It should instead summarize:
- The research question and the theoretical motivation for it.
- The planned design, sample size, and how that sample size was determined, such as a power analysis.
- The primary preregistered hypotheses and the planned analysis approach.
- A brief note that the protocol has received, or is undergoing, in principle acceptance.
Stage 2 Registered Report Abstract
Written after data collection and analysis are complete, once the protocol has full acceptance. It should include:
- The same background and design summary as the Stage 1 version, kept consistent with the accepted protocol.
- The actual results for each preregistered hypothesis, reported whether they were significant or null.
- A clear statement of any deviations from the original registered protocol, and why they occurred.
- A conclusion that interprets the confirmed, as opposed to merely planned, findings.
Frequently Asked Questions
How long should a research paper abstract be?
Most journal abstracts run 150 to 300 words; dissertation abstracts can run up to 350 words or one page. Always confirm the exact limit in the target journal’s or institution’s author guidelines, since limits vary.
What is the difference between an abstract and an introduction?
An abstract is a self-contained summary of the entire paper, including results and conclusions, read independently of the full text. An introduction is part of the paper itself and sets up the problem without revealing the findings.
Should an abstract include citations?
Generally no; most style guides and journals discourage in-text citations in abstracts because the abstract must stand alone, often in a database, without the paper’s reference list attached.
Can you write an abstract before finishing the paper?
A full paper should be finished, or nearly finished, before the abstract is written, since the abstract must accurately summarize the actual methods, results, and conclusions rather than planned ones. A conference abstract is the main exception; it is often submitted before the full study is complete.
What verb tense should an abstract be written in?
Use past tense for completed methods and results (“We measured,” “Scores increased”), and present tense for established facts or the paper’s ongoing conclusions (“These findings suggest,” “The model demonstrates”).
How do you write an abstract for a theoretical or non-empirical paper?
Replace the methods and results sections with a statement of the paper’s central argument, the framework or literature it engages, and its main theoretical contribution or conclusion, since there is no dataset or measured outcome to report.
What is a graphical abstract and do you need one?
A graphical abstract is a single image summarizing a study’s design or key finding, submitted alongside the written abstract, not instead of it. Some journals require one, many make it optional, and some do not use them at all, so check the specific journal’s submission guidelines. If a graphical abstract is optional, it’s still a good idea to submit one as a graphical abstract improves visibility of your article and boosts views, downloads, and cites.
Do all journals require structured abstracts?
No. Structured, subheaded abstracts are common in clinical, biomedical, and many social science journals, while humanities, some engineering venues, and many conferences still favor a single unstructured paragraph; the required format is always specified in the journal’s author guidelines.
This article was originally published on October 6, 2023, and updated on July 1, 2026.
Leave a Reply