Tuesday, August 25, 2009

Abstract Musings

How much effort do you put into writing scientific abstracts? Do you craft them carefully or treat them as an afterthought?

Here are some reasons to make an effort to write good abstracts.

First, the abstract may be the only part of a scientific article that is read. People often do not have access to the full article (requires a subscription or payment) or don’t want to spend the time reading the entire document, so they glean whatever they need from the abstract. Thus, it makes sense for your abstract to be as clear and concise as possible and to convey the essence of the paper and its significance. The more key words you include in your abstract, the more people are led to your paper and are likely to cite it.

Second, the abstract is often a preview of how good the paper is. Reviewers and editors form their first impressions of a submitted manuscript from the abstract. I know some reviewers who immediately reject a paper based solely on what they see in the abstract. Some editors (consciously or subconsciously) may use the abstract to decide whether to send out a paper for review. You might be surprised and outraged at this, but consider it from the reviewer’s viewpoint. Reviewers (and most associate editors) are providing their time for free and do not want to waste it on poorly-written manuscripts—ones that should be more polished before submission. The best way to annoy a reviewer is to submit a sloppy manuscript—and the abstract often announces just how bad it’s going to be.

The abstract tells the reader a number of things about the paper and the author(s) in addition to what the paper is about and what the main findings are. A poorly-written abstract says that the author is either a novice and does not understand how to write an abstract or is somewhat experienced, but did not put sufficient effort into the paper to write a proper abstract. Either way, it suggests that the rest of the paper is likely to be less than stellar. There are exceptions, of course, in which the paper turns out to be pretty good. However, I’ve seldom run into that situation.

When I get a paper to review, I first take a close look at the abstract. It tells me a lot about how much effort I will have to go to make a decision about it and to provide constructive criticism. I can often predict what the journal's decision will be based solely on the quality of the abstract. More specifically, I can spot the papers that will be accepted with minor revision. There will be some good papers with so-so abstracts; these are often the ones that are acceptable with major revision. The papers with atrocious abstracts often end up in the reject pile.

Note that as a reviewer/assoc. editor, I always read the ENTIRE paper and judge it on all its merits. The point is that the abstract sets the tone for the paper, which may take considerable positive points in the remaining narrative to overcome an initial bad impression. In contrast, a good abstract makes an immediate positive impression.

A bad abstract can also result when the paper is not really very interesting or important, and the abstract simply reflects this fact. This situation calls for even more effort on your part to prepare a good abstract. You must be creative in selling your paper, and the abstract is the advertisement for it. You can’t exaggerate, of course, but you should “put your best foot forward” in the abstract. It can make the difference between acceptance and rejection.

A really good abstract tells the editor that the work is important and interesting and that the author is a skilled scientist and writer. A “meaty” abstract contains lots of good information and creates a sense of curiosity in the reader, who then can’t wait to read the entire paper to get a full understanding of what was done, how it was done, and what it all means. You want your abstract also to say to the reviewer or editor that you are a professional (you’ve done your job and are not depending on the editor or reviewers to do it for you).

A final reason to spend time on your abstract is that it often helps you identify weaknesses in the paper that you then fix during later revisions. If you are having trouble summarizing your work in 300 words or so, then your paper may lack focus or organization or is unclear as to the work's significance.

Authors are often advised to write the abstract LAST, but this sometimes is translated by students as the abstract being the least important component of the paper. If you simply pull statements from the narrative and stick them together in a string, you are not really preparing a good summary of your paper.

So what makes a good abstract? A good place to start is the top journals in your field. Read and carefully analyze the abstracts of papers published in the best journals. They will all contain key elements. If you are a novice writer, a good formula to follow is:

1. Aim (state the problem or study objectives)
2. Location (briefly say where the study was conducted) or type of system, species, etc. you studied.
3. Methods (briefly describe the approach)
4. Results (concisely state the MAIN results—the ones that directly relate back to your stated objectives)
5. Main conclusions and significance (the “take-home message”)

Pay close attention to the word limit required by journals for abstracts. The shorter the abstract, the more difficult to write and the more effort required to get it right.

Practice writing abstracts. If you haven’t written any papers yet, select some good articles in a major journal and write an abstract for them based on your reading (don’t look at the abstract, though). Then compare your attempt with the published abstract.

So, if you are dreading writing that abstract, don’t procrastinate—get busy and develop your skills so that your next one sends the message: “Read this paper; it will be worth your while”.

Image Credit: "Abstract Thoughts" and oil painting by Aleta Gudelski

No comments: