An abstract is a summary of a document's main points that appears just after the title in most journals, providing readers with a means of assessing whether the article matches their interests. Since abstracts are increasingly being used in indexing, they may even appear on their own in indices of journals within a single subject area or in online databases. As a result, the abstract should (1) represent the full argument contained in your article: topic, reason for the study, methods used, results obtained, conclusions/implications from your study and (2) stand independent of your article.

Do not assume your readers have read your abstract prior to embarking on reading the article. In some cases, readers may skim your abstract in lieu of reading the article; in others, they may scan the abstract and promptly forget its contents as they begin reading the article itself (mostly because the information contained in the abstract becomes irrelevant once you read the article it represents.

Most journals specify a maximum length for abstracts, usually from 100-200 words, leaving its structure up to the author. Most commonly, writers tend to structure abstracts by introducing the topic in the present tense, usually in a sentence or two. They outline the background or need for the study in another few sentences without, however, referring to individual prior studies. Most writers then tend to accompany this section with several sentences describing methods and results in the past tense, providing major conclusions and implications stated in a sentence or two in the present tense.

Abstracts can, however, also fall into one of two categories: informative (described above) and descriptive or indicative, sometimes used for research reviews, and conference reports. Unlike informative abstracts, descriptive abstracts tend to provide little information about the study itself or its results, making it both less helpful to readers--and, not surprisingly, less common in research literature.

Like executive summaries, abstracts may be used by reviewers at the beginning of the review process to help officials sort proposals; reviewers may use it to preview the larger document; program officials may use the abstract to publicize the agency's funding decisions after grants have been awarded. As a result, the National Science Foundation (NSF) requires that proposal summaries should be "suitable for publication" and "informative to other persons working in the same or related fields and, insofar as possible, understandable to a scientifically or technically literate lay reader." Unlike research article abstracts, proposal abstracts run from 200-300 words and, if the agency specifies they may be longer, may be still more elaborate, like the NSF's revised proposal summary guidelines that specify abstracts or proposal summaries of up to one page in length.

When writing review/literature synthesis articles that often contain no single main point, you should describe only the general contents of the article rather than specific findings, writing the entire abstract in the present tense.
 

Informative science abstract:

A.J. Boulton, J.H. Bowker, M. Gadia, J. Lemerman, K. Caswell, J.S. Skyler, and J.M. Sosenko (1986). Use of Plaster Casts in the Management of Diabetic Neuropathic Foot Ulcers. Diabetes Care 9(2): 149-152.

Neuropathic foot ulceration is a major medical and economic problem among diabetic patients, and the traditional treatment involves bed rest with complete freedom from weight bearing. We have investigated the use of walking plaster casts in the management of seven diabetic patients with long-standing, chronic foot ulcers. Although all ulcers healed in a median time of 6 weeks, this therapy was not without side effects. We conclude that casting is a useful therapy for neuropathic ulcers, although several clinic visits, including cast removal and foot inspection, are necessary to avoid potential side effects caused by the casting of insensitive feet.

KEY WORDS: Surgical Casts, Foot Diseases, Diabetic Neuropathies, Skin Ulcer Therapy.
 

Humanities abstract:
J. Yellowlees Douglas, "Virtual Intimacyô and the Male Gaze Cubed: Interacting with Narratives on CD-ROM," Leonardo 29(3), 1996: 207-215.

Interactive narratives on CD-ROM have the potential to make us feel more a part of the stories we experience than do either film or print. Yet, as some early offerings in this fledgling genre reveal, the use of subjective camera and the constraints on the user's interaction with the narrative can conspire to make some readers feel straitjacketed by the text, rather than transported by it. The author analyzes one of these early offerings in depth and presents some alternative possibilities for this new genre.