The Art of the Academic Abstract

The abstract is the part of a paper most people will read, and for many of them it is the only part. Database searches return it and editors screen submissions by it, often before anyone opens the full text, while a committee member leafing through a dissertation's front matter forms a first judgment of the whole project from a page or less of prose. No other section carries as much weight per word.

What the abstract has to do

Whatever the discipline, a working abstract answers four questions in compressed form: what problem the study addresses and why it matters, how the author went about it, what the author found, and what those findings mean for the field. Purpose, method, findings, implication. A reader who finishes the abstract should be able to restate each of the four in a sentence of their own, and if any one is missing or vague, that reader has to open the paper to recover it — which a reviewer working through a stack of submissions may decline to do. Proportion matters as much as presence. Method and findings usually deserve the most space, since they are what distinguish this study from the many others asking similar questions.

The journal or institution decides the form. Some require a structured abstract, with labeled sections such as Background, Methods, Results, and Conclusions; the convention is common in the health sciences, nursing, and much of the social sciences, where readers scan for particular information and want to reach the results without reading the setup. Others, especially in the humanities, expect an unstructured abstract, a single paragraph that moves through the same logic without headings. The underlying tasks are identical. Labels only make the sequence visible.

Word limits come from the same authorities. Find the number before drafting. The journal's author guidelines or the graduate school's formatting manual will state it, and the submission system will often enforce it mechanically, cutting or rejecting text that runs over; because a dissertation abstract and an article abstract are usually governed by different counts, an abstract written for the dissertation will need to be cut and reshaped before it can accompany an article drawn from the same research.

Where abstracts go wrong

The most common fault is an opening that spends half the word count on background. Three or four sentences establish that the topic matters, that scholars have studied it, and that questions remain, and by the time the abstract reaches the study itself there is room for one sentence on method and a vague gesture toward results. One sentence of context is usually enough. The reader is already looking at this abstract because the topic interested them.

Missing results come next. An abstract that says the study "explores" a question and then stops has described an intention and reported nothing. State what the data showed or what the argument establishes, as plainly as the evidence permits, so that a reader who goes no further still leaves with the result.

The opposite fault is harder to see from inside the project. Authors sometimes claim more in the abstract than the paper supports: a small qualitative study described as demonstrating a general pattern, or a correlation reported in language that implies cause. Reviewers compare the abstract against the body, and a mismatch costs the author credibility that the rest of the paper then has to win back. Undefined abbreviations do quieter damage. Because an abstract is indexed and read apart from the paper, it has to stand alone, so any acronym used there needs its full form at first mention even when the paper defines it on page two, and a specialized abbreviation that appears only once is better spelled out entirely.

Write the abstract last, or at least revise it last. A draft written at the proposal stage describes the study the author planned; the finished paper reports the study the author conducted, and the two drift apart in ways that are easy to miss once the research is underway — a narrowed question, say, or a result that complicated the hypothesis. When the paper is complete, reread the introduction and the conclusion, then rebuild the abstract from what those sections now say. Check it against the word limit and against the body. Every claim in it should point to a page.

If you would like a second reader for the abstract and the paper behind it, OTG offers journal article editing for manuscripts headed to submission.