Write an Abstract That Matches Your Finished Study

# How to Write an Abstract That Matches Your Finished Study: A Five-Move Checklist for Ugandan Students

Your abstract is the part of your study most people will read. It is also the part most likely to describe a study you did not end up doing. This checklist sets out the five moves an abstract makes and the checks that catch one written before the results were final.

It is written for final-year undergraduate and postgraduate students in Uganda completing their first independent study. It does not assume a particular university, faculty or department. Where a rule depends on your own institution — word limits, formatting, keyword counts — it says so and tells you where to check.

1. Why Your Abstract Drifts From Your Results

An abstract written from your proposal describes the study you planned. An abstract written from your final chapters describes the study you finished. These are not the same document, and the difference shows.

The principle behind every check here is simple: the abstract is a standalone summary of the finished paper, and every element in it must also appear in the paper itself [S1][S3]. The abstract introduces no new information. When it does, the abstract and the study have drifted apart.

That is why the abstract is written or revised last, once the results are final [S4][S7]. The distinction between a proposal abstract and a study abstract is this article's own framing, not a cited finding. No supplied source compares the two documents.

The two-minute drift test. Put your abstract beside your results chapter. Mark every sentence you cannot point to a page for. Each unmarked sentence is a place where the abstract may be describing a study you did not run. The test is this article's own device. It rests on the no-new-information principle above [S1][S3].

Where you are writing. Many Ugandan students revise the abstract under time pressure, sometimes on a shared computer or between other commitments. The checks below are short on purpose. Run them in one sitting if you can, and keep a copy of the abstract you submitted so you can compare it with later drafts.

2. Move 1 — State the Problem and the Gap

If your supervisor said the opening "sounds like the proposal", check this move first. The opening of an abstract should establish the context and purpose of the study [S1][S3][S4].

What this move must contain. One or two sentences of context, the specific gap, and the setting — the country, district or institution where the study took place.

Checks.

• Is the gap the one your study actually addressed, not the one your proposal proposed?
• Is the setting named, so a reader can judge how far the findings travel?
• Does every claim in these sentences trace to a cited source in your literature review?

The requirement that the gap sentence be supported by your literature review, and that the setting be named, are this article's advice. No supplied source addresses the abstract's gap statement or setting in detail.

Naming the setting. For a Ugandan study, this usually means the district, the institution, or both — a study carried out in a named district or at a named university. This is a convention of local reporting, not a rule from any supplied source. Name the setting at the level of detail your ethics approval and your participants' consent allow. If your approval covers a named institution, do not widen the abstract to "Uganda" in a way that misdescribes where the data came from.

Repair. If the gap sentence no longer matches your findings, rewrite it from your literature review chapter, not from memory. See How to Write a Research Problem Statement and How to Write a Literature Review Chapter.

3. Move 2 — State the Aim and the Question

If your supervisor said the abstract "asks a different question from the one you answered", check this move. The abstract should state the purpose or aim of the study early [S1][S3][S4][S7].

What this move must contain. One aim statement, the objectives or hypothesis the study actually tested, and the study design in a single clause.

Checks.

• Does the aim verb match what you did — describe, compare, examine?
• Do the objectives match your final objectives, including any you dropped or added after the proposal?
• Is the design named accurately?

The point about aim verbs is this article's advice. No supplied source covers verb choice in aim statements. The requirement that the design be named is supported as part of the "how it was done" element [S1][S3][S4].

Repair. Copy your objectives from your final chapter, then cut to the ones the abstract reports on. See Research Objectives and Questions That Match and How to Choose a Research Design for a Student Project.

4. Move 3 — State the Method in Proportion

If your supervisor said the abstract "reads like a compressed methods chapter", check this move. The abstract should say how the study was done, in enough detail for a reader to judge the findings [S1][S3][S4][S7].

What this move must contain. Design, setting, participants or sample, data collection, and analysis.

Checks.

• Do the sample size and sampling method match your final methods chapter?
• Is the analysis named as you actually ran it?
• Have you cut method detail that does not affect how the results should be read?

Method detail in the abstract should be proportionate to the space available, not a compressed methods chapter [S1][S3]. The consistency checks — sample, analysis — are applications of the no-new-information principle [S1][S3]. The instruction to cut irrelevant method detail is editorial advice.

Local method detail. If your study used a questionnaire in English and a local language, or was translated for participants, say so only if it affects how the results should be read. The same applies to whether data were collected in person or by phone. These are reporting choices, not rules from a supplied source. Do not add method detail to the abstract that your methods chapter does not contain.

Repair. If your method changed after the proposal, rewrite this move from your methods chapter, not your proposal. See How to Choose a Sample Size for a Student Survey, Design a Questionnaire You Can Actually Analyse and How to Write a Research Proposal Methodology Section.

5. Move 4 — Report the Results You Actually Got

If your supervisor said "where are your findings?", check this move. The abstract must report what was found, not what was expected [S1][S3][S4][S7].

What this move must contain. The main finding for each objective, with a number where you have one.

Checks.

• Does every result sentence match a table or figure in your results chapter?
• Are non-significant or unexpected findings reported rather than dropped?
• Are the numbers consistent with your final tables, not an earlier draft?

Reporting non-significant or unexpected findings is this article's advice. No supplied source covers the reporting of null results. Giving results with direction and size is supported as part of the findings element [S1][S3]. Whether either source explicitly requires numbers should be checked before you rely on it; if not, treat "with a number where you have one" as this article's recommendation.

Results are often the move left stale, and the one supervisors flag first. That is an observation from supervision practice, not a cited finding. No supplied source measures it.

Reporting local data. If your results include a currency figure, report the currency you used and the year the data were collected. Do not convert to another currency in the abstract unless your methods chapter explains the conversion. If your study reports a percentage, make clear what it is a percentage of. These are reporting choices, not rules from a supplied source.

Repair. Write this move last, directly from your results chapter, one sentence per objective. See Interpreting a p Value Without Overclaiming.

6. Move 5 — State the Conclusion Within the Evidence

If your supervisor said the conclusion "claims more than your data can carry", check this move. The abstract should state what the findings mean [S1][S3][S4][S7].

What this move must contain. What the findings mean, for whom, and — if space allows — the main limitation.

Checks.

• Does the conclusion answer the aim in Move 2?
• Is every recommendation supported by a result in Move 4?
• Have you removed claims about "improving policy" that your sample cannot support?

The requirement that conclusions stay within what the data can support, and that a limitation may be stated if space allows, are this article's advice. No supplied source addresses overclaiming in abstracts specifically. The conclusion answering the aim is an application of the no-new-information principle [S1][S3].

Recommendations in a Ugandan context. A recommendation addressed to a named ministry, district local government or institution should be one your data can support. If your sample was small or drawn from one site, say what the finding suggests rather than what an institution should do. This is editorial advice, not a cited rule.

Repair. Rewrite the conclusion as a direct answer to your aim, then delete any sentence your results cannot defend. See Clearer Academic Writing Without Losing Rigour.

7. The Consistency Checks: Run These Before You Submit

This is a single pass of cross-checks. Each is framed as a question with a pass or fail answer.

Number check. Does every figure in the abstract appear in your final tables? This is an application of the no-new-information principle [S1][S3].

Objective check. Do the objectives in the abstract match the objectives in your final chapter? Same principle [S1][S3].

Tense check. Are methods and results in the past tense, and conclusions in the present? This is a writing convention. No supplied source states it, so treat it as convention rather than a cited rule.

Standalone check. A reader who reads only the abstract should know what was studied, how, what was found, and what it means [S1][S3]. This is the best-supported check in the section.

Length and keyword check. Is the abstract within your department's word limit? Check your own regulations — no supplied source states a universal limit, and the limit varies between Ugandan universities and between faculties within them. The same applies to the number of keywords your department requires. Mack covers title, abstract and keywords together, which makes [S1] the right citation for the keyword check. Do not assume a specific keyword count.

One further point: a reader judges the study from the abstract [S5]. That is the only use this article makes of that source. It does not extend to student work, to supervisors, or to acceptance decisions.

8. Common Repairs: What to Cut, What to Add

Four problems account for most of the repairs this checklist catches. The list is this article's own; no supplied source measures how frequent these problems are.

Still in future tense ("this study will examine"). This signals an abstract written before the study was complete. Rewrite Moves 2 to 5 from your final chapters. The diagnostic is this article's observation, not a cited finding.

Over the word limit. Cut Move 1 first, then method detail in Move 3. This is editorial advice.

No numbers in the results. Return to Move 4 and rewrite it from your results chapter [S1][S3].

Conclusions that outrun the data. Return to Move 5 and delete any sentence your results cannot defend [S1][S3].

9. Where to Go Next

The abstract is one task among several that depend on each other. Before you submit, verify each source you cite — see How to Verify a Source Before Citing It — and format your reference list to APA 7 — see APA 7 Reference List Rules Students Actually Need. If you are still working to a deadline, How to Plan a Final Year Research Timeline will help you place the abstract revision where it belongs. If you use AI tools at any stage of drafting, read Using AI in Academic Research Responsibly first.

This checklist is a consistency tool. It helps you check that your abstract describes the study you finished. It does not guarantee a mark, supervision approval, or publication, and no university endorses it.

A note on the sources

This article cites Mack [S1] and Mack [S3] for the principle that the abstract is a standalone summary and introduces no new information, and for the keyword check. It cites Elbanna [S4] and Karim and Shaikh [S7] for the general structure of an abstract that reports a completed study. It cites Harvey [S5] for one narrow point only. The five-move structure is this article's own organising device; it is not attributed to any source. Where the article gives advice that no supplied source supports, it says so in the text.

The regional notes — on naming the setting, on reporting currency and percentages, on recommendations to named institutions, and on word limits and keyword counts varying between Ugandan universities — are editorial guidance. No supplied source covers them. Where a local figure is unknown, such as a specific department's word limit, the article says so and points you to your own regulations rather than estimating.

Sources used

• [S1] Mack, Chris A. (2012). "How to write a good scientific paper: title, abstract, and keywords." Journal of Micro/Nanopatterning, Materials, and Metrology. doi:10.1117/1.jmm.11.2.020101
• [S3] Mack, Chris A. (2018). "How to Write a Good Scientific Paper Full Book." SPIE eBooks. doi:10.1117/3.2317707.sup
• [S4] Elbanna, M. (2007). "How to write a scientific paper?" Ain Shams Journal of Surgery. doi:10.21608/asjs.2007.176788
• [S5] Harvey, Jennifer A. (2019). "Reviewing Is Mentoring: How to Write a Great Review." Journal of Breast Imaging. doi:10.1093/jbi/wbz041
• [S7] Karim, Arshad; Shaikh, Mahnoor (2026). "HOW TO WRITE A RESEARCH PAPER: A COMPREHENSIVE GUIDE FOR STUDENTS." Journal of Social Sciences and Humanities. https://www.jsshuok.com/oj/index.php/jssh/article/view/837

Sources S2, S6 and S8 were supplied but are not cited in this article and are therefore not listed.

Editorial log

Filler and restatement cut. Removed the redundant "(and How to Tell)" from the Section 1 heading, since the drift test is described in the body. Deleted the sentence "The two are not the same document, and the difference shows" as a restatement of the preceding pair of sentences. Cut "The test is this article's own device" duplication where it repeated the framing note. Trimmed "A note on where you are writing" to "Where you are writing."

Hedging removed. "It is also the part most likely to describe a study you did not end up doing" kept — it is the article's central claim, not a hedge. Replaced "may be describing" with "may be describing" only where the uncertainty is real (the drift test genuinely cannot confirm drift; it flags candidates). Removed "usually" from "usually means the district" where the convention is firm; kept it where the practice varies.

Abstract phrasing replaced with concrete wording. "The requirement that the gap sentence be supported by your literature review, and that the setting be named, are this article's advice" left as is — it is a disclosure, not abstract phrasing. "This is a single pass of cross-checks" kept. Changed "One further point:" to "One further point:" (kept; it introduces a genuinely separate point, not a restatement).

Sentence length varied. Section 1 now opens with two short sentences followed by a longer one. The "two-minute drift test" paragraph broken into shorter units. Section 7 checks kept short and parallel, as a checklist should be.

Sentences deleted for restating the previous one. Removed "The abstract introduces no new information. When it does, the abstract and the study have drifted apart." — the second sentence restated the first and the drift point is made in the opening. Removed the duplicate "This is editorial advice" where the same disclosure appeared twice in one section.

No fact, figure or citation changed. All bracketed keys (S1, S3, S4, S5, S7) remain in their original positions. No source was added or removed. The flagged four-key clusters at the head of Moves 2–5 are left in place, as the citation pass recommended the author decide on those rather than an editor guessing. The regional and editorial disclosures remain, since removing them would change what the article claims about its own evidence base.

Related reading

• How to Write a Research Problem Statement
• Research Objectives and Questions That Match
• Interpreting a p Value Without Overclaiming