Scientific Abstracts: Explained!

Scientific Abstracts: Explained!

Science Writing

In theory, abstracts are supposed to have 5 parts. In practice, however, they are structured much differently.

Table of contents

Scientific abstracts give your busy reader the information she needs in order to decide whether to read your report.

Abstracts are made up of the answers to 3 questions:

  1. What’s the problem?
  2. How did you solve it?
  3. What did you find?

A common mistake is to include too much information, which can make it hard for the Reader to find the answer to these questions.

The goal of an abstract is not to condense all the information from the report into 5-6 sentences. Our goal is to make the Reader read the rest of our report.

Let's get started.

In this guide you’ll learn:

  • How to start a scientific abstract;
  • How to identify the 5 parts of a scientific abstract;
  • How to choose the structure that’s best for your scientific abstract.

🤓 How to Start a Scientific Abstract

The way to begin your scientific abstract is not to write. It’s to imagine your Reader.

If we want the Reader to read our report, then we need to give the Reader what she wants.

Who is the Reader?

  • Your Reader is not your teacher. Teachers get paid to read your writing, no matter the quality. But real readers are totally different. There is nothing stopping real readers from closing the book and turning on YouTube. Make your writing valuable by solving a reader problem.
  • The Reader is a smart person. She just doesn't know what you know. She might not be a specialist, but she has a general sense of the field - its problems, methods, discourse conventions, etc. Write to her, not to a general audience.
  • The Reader is a busy person. There are many things she would rather be doing: conducting her own research or teaching, being with friends and family, hobbies, sleep - anything other than reading scientific reports. This is not fun for her; this is work. Make it easy for her to find what she wants.
  • The Reader has Instagram. Her attention is being pulled away every moment by an algorithm that wants to give her pictures of beautiful people and cute puppies. The competition is fierce. You need to capture this attention and sustain it. Don't just give her information; raise her curiosity. Write in a way that makes her want to find out more.

What does the Reader Want?

  • The Reader wants novelty. After scanning through countless reports, she's tired of clinical language and mundane findings. Make it obvious that your research provides something new.
  • The Reader wants to move quickly. The Reader skims, concentrating on particular passages and skipping over anything she finds invalid or irrelevant. Provide the information she wants, nothing more.

The key to writing a good scientific abstract is to keep the Reader in mind.

🖐 What Are the 5 Parts of a Scientific Abstract?

The idea that scientific abstracts have 5 parts is widespread, but it's also wrong.

The common framework given for scientific abstracts is

  1. Introduction (I) - Establish the context and motivation of the research.
  2. Purpose (P) - State the purpose, thesis, or hypothesis.
  3. Method (M) - Describe the design, procedures, assumptions, approach, data, etc.
  4. Product (Pr) - State the main findings or accomplishments.
  5. Conclusion (C) - Interpret or extend the results beyond the scope of the paper; draw inferences; point to wider implications or applications.

These 5 parts of the scientific abstract correspond roughly to the overall organization of the scientific report itself: Introduction - Methods - Results & Discussion - Conclusion.

The problem with this popular framework is that, in practice, scientific abstracts aren't actually written that way.

In fact, less than 5 percent of published academic papers contain all five parts. In Disciplinary Discourses (The University of Michigan Press: 2004), Ken Hyland analyzes the abstracts from over 800 published academic articles. Here's what he found:

  • 50% of the papers contained no Method section;
  • 55% omitted the Introduction;
  • Only 22% included a conclusion.

So let's forget the ideal of the 5-part scientific abstract. How are scientific abstracts actually written?

  • 94% included a Product (Pr) statement;
  • 25% featured a Purpose-Method-Product (P-M-Pr) schema;
  • 15% featured an Introduction-Purpose-Product (I-P-Pr) schema.
Key Hyland, Disciplinary Discourses (pp. 70)

The 5-part abstract is a myth.

🤔 Which Structure Is Best for Your Scientific Abstract?

As the chart above shows, you have some options when it comes to structuring your scientific abstract. This raises some interesting questions:

  • Should my abstract contain an Introduction? A conclusion?
  • Do I need to provide background information in my abstract?
  • Which abstract structure is best?

To answer these questions, consider the field that you're writing in and the community that you're writing to.

Citing Stephen Toulmin's suggestion in Human Understanding (Princeton University Press: 1972) that research disciplines can be arranged along a continuum from "compact" to "diffuse," Hyland argues that research communities in the sciences are urban (i.e., 'compact') and those in the soft sciences and the humanities are more rural.

Writers in urban research communities use one kind of abstract, while those in rural communities use another.
In general, research communities in the hard sciences are 'urban,' and those in the soft sciences & humanities are more 'rural.'

Abstracts for the Hard Sciences

The hard sciences, according to Hyland, are characterized by "close-knit urban fields." In these fields,

a small, discrete number of problems have relatively large numbers of people working on them. Researchable problems are therefore fairly well defined and there is a general expectation that readers will be familiar with the issues in which any piece of research is embedded.

For readers in the hard sciences, you don't have to spend a lot of time explaining the background and methods. There's a lot of shared knowledge, so it's safe to assume that they already possess this information.

The most common abstract structure in the hard sciences is P-M-Pr because these readers basically just want you to answer 3 questions:

  • What's the problem? (Purpose)
  • How did you solve it? (Method)
  • What did you find? (Product)

Abstracts for the Soft Sciences and Humanities

In the soft sciences and humanities, on the other hand, there's not really such a well-defined set of problems - researchers are more spread out, working on their own topics of interest.

A marketing researcher interviewed by Hyland captures this idea perfectly:

There are just so many live topics in my field. No-one can keep abreast of them all so it's necessary to establish, or at least reestablish anyway, the importance of the subject.

Nor is there a well-established method of solving problems in rural research communities. Researchers vary in their approach, depending on their background and discipline.

Consequently, Hyland notes, "writers have to work much harder to acquaint readers with the background to their research."

Because research in the humanities and social sciences tends to be more diverse and have more permeable boundaries, statements which functioned to provide a general context were more common.

Abstracts in the soft sciences, then, tend to...

  • feature lengthy introductions that provide context for the research;
  • omit the methods section.

🥳 In Sum

When you're writing your abstract, try to keep these points in mind:

  • The Reader is an intelligent peer who is in a hurry. Write to her, not to a general audience.
  • The goal of your abstract is to get the Reader to read your report. Do this by answering the Reader's questions as quickly and clearly as possible.
  • Different readers have different questions, depending on the research community that they belong to.
  • You don't need to include all 5 parts of the abstract (most writers don't!)

More posts in Science Writing