While the abstract started as a summarisation tool, nowadays, search engines and electronic bibliographies index and catalogue titles, abstracts and keywords, allowing your work to be easily discovered, read, used and cited by scholars who may not otherwise have been aware of it.

An abstract must be well written, clear, specific, coherent, effective and powerful. It should communicate a manuscript’s scope to as wide an audience as possible. If it doesn’t do its job, readers “may turn to other abstracts that do” (Foster). The limited word count means that every word should be essential.

I pay close attention to abstracts, providing suggestions about what may be further included and/or implied therein. Effectively encapsulating a body of research in an abstract usually takes repeated rewriting. However, with time pressure always a factor, the risk is that the abstract is rushed.

The abstract’s purpose
Given the time and effort that goes into delivering reliable, rigorous research, it would be a pity to drop the ball in the abstract, since – besides the title – this is the most read section of an article, thesis or dissertation. Abstracts also appear in invitations to talks and lectures as well as in conference proceedings. Yet they seldom represent their body text 100%.

An abstract gives readers the gist of a manuscript quickly, so that they can decide whether they want to read it (or attend your talk). It also prepares readers to follow the information in the body text. Identifying critical assumptions, non-standard methods, and limitations on the findings will make clear a research project’s scope and potential applications.

Guidelines vary
Guidelines for maximum word counts and what to include in an abstract vary between paper types, fields, journals, conferences and universities. Some now even allow for a highlights element and/or a graphic or infographic element.

Many disciplines have specific keyword lists that may be helpful to an author. Keywords should not be too general. Journals tend to want around 3 to 6 keywords; these should also appear in the body text.

IMRaD and beyond
How much of an abstract you devote to each information type or section and the sequence vary, depending for instance on a paper’s nature and genre. This calls for “a balance between including detail and retaining the reader’s interest” (Foster). Information is sometimes implied rather than stated. Well-placed adjectives (e.g. “integrative”) can be very helpful, especially when the word count limit looms.

Regarding IMRaD (introduction, method, results and discussion) structure, I recommend the guidance of the OWL at Purdue and the University of Wisconsin.

Some tips and best practices
• Follow the guidelines of the institution you’re submitting to.
• Keep the abstract to one paragraph, unless you have good reason to do otherwise.
• Understand your audiences (are they general or specialised?) and what their members know.
• If you use abbreviations in the abstract, unpack them there.
• Don’t mislead; don’t overstate certainty; acknowledge when findings are preliminary.
• An abstract is not an advertisement.
• Read other abstracts, and save the two or three most striking ones (not necessarily from your topic or discipline) to refer to when you write an abstract.
• Write or rewrite the abstract after signing off your paper; make time for this.
• Ask a colleague for incisive feedback to your abstract. Disclose the word count limit.
• Have at least the abstract, introduction and conclusion (if not the whole document) professionally copyedited.

What to avoid
Avoid: extensively referring to other studies; defining any terms (at length); adding new information that doesn’t appear in the body text; adding filler words and obscure jargon; including references. Introducing citations undermines the expectation that the abstract must stand alone; it also blocks readers’ flow.

Don’t include personal research plans, except in an internal report or an application for a research grant. Also, don’t undermine the article’s authority.

Some abstracts summarise a paper rather than its content. Such meta-reporting may lead to “uninformative boiler-plate text” (Foster). Don’t write: “Implications of the results are discussed.”

Present tense, first person
While some journals used to require that an abstract be written in the third person (e.g. “the author analyses”) and in the past tense, many increasingly require the present tense and the active, first-person perspective. Consult the guidelines.

Some journals withhold information for commercial gain
Regionel addresses a problem in some journals: “reading research abstracts that tell very little of the salient findings of the paper, particularly those behind a paywall, causes frustration… those abstracts work more as a marketing strategy rather than to disseminate important information… For publicly-funded researches… withholding information for commercial gain, appears to be unethical or defeats the purpose of research”.

Open-access publishing has gained popularity. There are many credible OA journals. While the publication costs are not universally high in all OA journals, many prominent ones have significant article processing charges, creating barriers, though some offer waivers or have very low fees. Institutional funding helps to cover one’s costs.

Predatory publishing has increased; please be careful. This may lead you to re-think who you want to submit to.

For more about abstracts (and my sources), go here.