Elements of Science: What are the elements of a great abstract?

The painful truth: no matter how fantastic your research is, there is a good chance that most people who come across it will only read your abstract. Ouch.

“People scan abstracts when they’re looking for papers to read,” says Dr. Cecile Badenhorst, Assistant Professor at Memorial University of Newfoundland and Labrador and author of several books on academic writing. “Readers will skip over articles if the abstracts are not attractive to them.” It’s therefore important that your abstract contains a clear, focused description of your study that highlights the information most relevant to readers.

Then there is the issue of access. “Abstracts are the only reliably open-access component of most peer-reviewed science right now,” says Dr. Bethann Merkle, a subject editor (science communication) for FACETS. “Researchers should view abstracts as one of the leading edges of open science initiatives and a critical part of efforts to do and share science that are more broadly accessible, useable… and hopefully more inclusive.”

With abstracts playing such an important role in promoting the discoverability and accessibility of your research, it’s worth giving them extra attention. Here are some tips that will help you create a great abstract.

Short yet comprehensive

Abstracts tend to be limited to a length between 200 and 300 words. “It’s not possible for an abstract to convey all the nuance and significance of a paper,” says Merkle. “But it should still provide genuine insight, even for people who don’t have ready access to [the rest of the paper] due to journal paywalls.”

Badenhost notes that abstract conventions vary by discipline and journal, but in general, the key elements of an abstract are a summary of the research problem, the purpose of the research, the methodology, the key findings and, lastly, a key message.

In this abstract from Nugroho et al.’s FACETS 2023 paper, “A combination of optimized threshold and deep learning-based approach to improve malaria detection and segmentation on PlasmoID dataset,” the summary of the research problem (yellow), the purpose of the research (green), the methodology (blue), the key findings (pink) and key message (red) are easily identifiable.

The research problem answers the question, “Why do I have to listen to this research? Why do I have to read it? It’s very important… because it provides the real-world rationale [for the study],” says Badenhorst. A research problem often arises from a knowledge gap. As such, you should also identify the knowledge gap your research aims to fill. “This satisfies the academic premise that all research builds on previous research and provides the academic rationale for this study. It also situates your research within a body of literature,” explains Badenhorst.

Next, add a sentence or two stating the purpose of this research. This statement should address the knowledge gap and introduce the broad goals of the study. You can also include a word that indicates the methodology used. “For example, if you write that the aim of your study is to ‘explore’ a phenomenon, it indicates that it’s a qualitative, open-ended type of study,” says Badenhorst. “Other methodology descriptors that might suit your study include ‘observe’, ‘quantify’, ‘analyze’, or ‘describe’.”

After you’ve explained the purpose of your study, include a brief summary of the methods and materials. You only need a few sentences; you’ll expand at greater length in the Methodology section of your paper. Similarly, you probably won’t have space to include all your results in the abstract. Instead, Badenhorst recommends highlighting one key finding. If you do have room to expand, focus only on what’s most important.

The final part of the abstract should be dedicated to a key message. “If you wanted someone to cite you on this paper, what would they say?” Badenhorst asks.

Think beyond English

English is the dominant language in research publications, however not every researcher or reader of research necessarily speaks or is very comfortable in English. As a result, language can be a huge barrier to the communication of scientific research. Merkle would like to see researchers publish abstracts in several languages, even if the manuscript is in English. For example, Khelifa et al.’s 2022 FACETS paper, “Language of citation and publishing performance of graduate students in French-speaking countries with different economic and linguistic advantages,” has abstracts in English and French—Canada’s two official languages. In this way, even though the paper itself is written in English, French readers have easy access to a summary of the study and its key findings.

There are, however, other languages prominent in Canada, which is why Merkle would vote for tri-lingual abstracts as the bare minimum. “One of those languages needs to be the language local to the region where the research was conducted,” Merkle explains.

For example, the abstracts for all the papers featured in the Arctic Science special issue, Knowledge Mobilization on Co-Management, Co-Production of Knowledge, and Community-Based Monitoring to Support Effective Wildlife Resource Decision Making and Inuit Self-Determination, have been translated into an appropriate Indigenous language.

Recognize your wider audience

One of the biggest mistakes Merkle sees is researchers having tunnel vision: conceptualizing, framing, and writing the abstract as a synopsis only understandable by specialists within the author’s field.

“The number one thing researchers should do is recognize that abstracts are tools used by people far beyond the colleagues in our specialized subfields,” she explains. To ensure that your abstract is accessible to researchers in distant fields, policymakers, and non-academics, one of the things Merkle recommends is writing abstracts in more accessible language.

Bandenhost suggests that writing in an active voice can help with both accessibility and engagement. “Passive language can be very awkward, dense, and long-winded,” she says, offering the example of the sentence we studied fifteen patients (active voice) versus “fifteen patients were studied” (passive voice). In the former example, it is clear to the reader who is doing the action of studying and can be understood more quickly. For similar reasons, Bandenhost also recommends avoiding acronyms, abbreviations, and jargon where possible, and always defining them on first use.

Some publishers encourage authors to submit a plain language summary as well as an abstract to help facilitate engagement and understanding. Regardless, having an abstract that is accessible to a broad audience will always be a boon.

Finally, don’t expect to be able to write a great abstract straight off. Creating an informative, engaging, and accessible abstract takes practice. Even seasoned researchers will revise their abstracts several times to get them right. By honing your skills, your abstracts will entice more readers to delve deeper into your work.

Tips, techniques, and expert advice on crafting great scholarly communications can be found in our Elements of Science blog series. Read about the elements of a great map, create a great briefing note or table, and learn how to write a great title.