How to Write an Abstract

By Marco Bomfoco (Guest blogger)

An academic abstract is a short restatement of all essential points of a research paper. The abstract is one single paragraph and is subject to specific word limits, typically under 300 words. It stands alone bellow the title or at the end of the paper. Note that an abstract is NOT an introduction or a plan to the paper. In the words of Craig W. Allin, "abstracts are an exercise in writing with precision and efficiency."

In fact, the abstract is written after the investigation and the whole article is completed. It should be written in the same language as the paper and should be translated into one of the world languages. We can say that the primary purpose of an abstract is to permit a quick appraise of the applicability, importance and validity of a research paper. But always recall that the reader KNOW the subject but HAS NOT READ the paper.

The abstract presents the information in four general sections: INTRODUCTION, METHODS, RESULTS and CONCLUSIONS. It is worth noting that an abstract is only text and follows strictly the logical order of the paper. That is, the abstract ought to parallel the structure of the original paper. At the same time, it adds NO new information, i. e. that is not stated in the paper. Now notice that the abstract can be viewed as an independent document. It is because of this that it should be unified, coherent (i.e. providing appropriate transitions or logical linkage between the information included), concise, and able to stand alone. In other words, the abstract should be complete in itself.

Surely, it is sometimes the case that an abstract will be read along with the title and in general it will likely be read without the rest of the document. In fact, we might consider that the abstract is the most important part of a scientific paper. It follows, then, that it is an absolute must to include all the keywords related to the study. Notice that keywords (also called search terms) represent the most important terms or concepts (words or phrases) relevant to your topic.

There are two types of abstracts: descriptive and informative. The descriptive or indicative abstract, identifies the contents of the research or the basic subject of the article, demonstrating the paper’s organization without providing results or conclusions. Thus, it is not very informative. This type of abstract is always very short, usually under 100 words; and it is useful for a long report. On the other hand, the informative abstract, which is also known simply as a summary, gives the principal argument and summarizes the principal data, providing the reader with an overview of the objectives, methods, results and conclusions of the study. So, be specific. You may also have heard of a "structured abstract" -- this is a subtype of the informative abstract which has more than one paragraph.

What to include?

The content of the abstract includes:

• Motivation and purpose: main subject or research question and review of the relevant literature.
• Specifics: problem statement, approach, objectives, hypothesis, research methodology (method(s) adopted or search strategies).
• Results: main findings (proposed solutions to the problem) and discussion.
• Conclusions and implications/outcomes: what the results mean and further points

As we can see, the abstract must state:

• The problem addressed and some background information.
• The solution or insight proposed (newly observed facts).
• An example that shows how it works.
• An evaluation: a comparison with existing answers/techniques.

Then, an abstract should provide answers for the following questions:

• What and why.
  
• What you found.
  
• How you did it.

But how do we begin?

What would be an effective way to begin an abstract? To help you on your way let us consider some introductory sentences.

First, let us see some opening sentences that DO NOT offer real information:

1. This paper reports on a method for...


2. The paper explores the notions of...


3. The purpose of our research is to consider how...


4. The objective of this study is to determine...

Thus, it is clear that you should avoid writing a statement of scope.

On the other hand, the sentences bellow represent good examples of introductory statements, for they go directly into the subject. They give something to the reader. Let us see how it works:

1. The development process of hypermedia and web systems poses very specific problems that do not appear in other software applications, such as…


2. Given a large set of data, a common data mining problem is to extract the frequent patterns occurring in this set.


3. According to many recent studies the effect of learning style on academic performance has been found to be significant and mismatch between teaching and learning styles causes learning failure and frustration.

Do’s and don’ts of abstract writing

• Do write a single paragraph.


• Do meet the specific word length.


• Do answer the questions: what, why, and how.


• Do use familiar language to the reader.


• Do use a few keywords.


• Do write short sentences.


• Do improve transitions between the sentences.


• Do use active voice.


• Do use third person singular.


• Do begin with a clear introductory statement written in the present tense.


• Do use past tense in the main body.


• Do write a concluding statement in the present tense: just tells what the results mean (e.g. "These results suggest...").


• Do fix grammar.


• Do use headings, subheadings and tables as a guide for writing.


• Do print and reread the abstract.


• Don't cite the sections of the paper.


• Don’t include references to the literature and to figures and tables.


• Don’t use abbreviations.


• Don’t add new information.


• Don't add superfluous information.


• Don’t add opinions.


• Don’t repeat information.


• Don’t repeat the article title.

Marco Antônio Bomfoco is a professor at the Faculty of Technology SENAC in South Brazil. A Ph.D. from Pontifical Catholic University of Rio Grande do Sul, and an M.A. from the same university, Marco is a long-time language enthusiast. His many interests include writing, general linguistics, anthropological linguistics, case theory and grammatical relations, history of linguistics, artificial languages, phenomenology, story telling, and myth. Article source here.