After the title, the abstract is the most-read part of an article. Editors use it to decide whether to send the paper out for review, and readers to decide whether the rest is worth reading. It must make sense without the rest of the article.
What it should contain
Most social science and health journals ask for a structured abstract in four parts, a format readers find more informative and easier to search (Hartley, 2004):
- ObjectiveWhat did you want to find out?One sentence, in the past tense.
- MethodHow did you do it?Design, participants (N, age, % women), instruments and analyses.
- ResultsWhat did you find?The main findings with their values: at least two numbers.
- ConclusionsWhat does it mean?The main implication, in the present tense.
How it should sound
- Self-contained. No citations, no undefined acronyms and no references to tables or figures.
- With data. “A relationship was found” says nothing; “r = .42” does.
- Faithful to the article. Only what is in the text: no new results and no conclusions stronger than the data (Tullu, 2019).
- Consistent with the design. As in the title, an association is not a cause.
Before and after
| Weak version | Problem | Improved version |
|---|---|---|
| Stress in university students was studied. | Vague aim, no variables or relationship | The aim was to determine the relationship between academic stress and performance in university students. |
| The results were significant. | No data to back it up | Academic stress was negatively associated with grade point average, r = −.31, p < .001. |
| Stress is shown to cause poor performance. | Causal conclusion in a cross-sectional study | The findings suggest that reducing academic stress could support performance. |
Two languages and keywords
If the journal asks for an abstract in Spanish and in English, the numbers must be identical in both. Keywords are best taken from a thesaurus such as MeSH, DeCS or the APA Thesaurus: those are the terms databases use to index articles and other researchers use to search.
What each journal decides. Maximum length, whether the abstract is structured or a single paragraph, the labels for each part and the number of keywords change from one journal or conference to another. Check their author guidelines.
Before you call it done
- Does it say what was done, how, what was found and what it means?
- Does it include at least two results with their values?
- Can it be understood without the article, with no citations or undefined acronyms?
- Is everything it says in the article, with no conclusions stronger than the data?
- Do the numbers match in both language versions?
The Manuscript Diagnosis checks that the abstract is complete and that its numbers match those of the abstract in the other language.
References
Hartley, J. (2004). Current findings from research on structured abstracts. Journal of the Medical Library Association, 92(3), 368–371.
Tullu, M. S. (2019). Writing the title and abstract for a research paper: Being concise, precise, and meticulous is the key. Saudi Journal of Anaesthesia, 13(Suppl. 1), S12–S17. https://doi.org/10.4103/sja.SJA_685_18
von Elm, E., Altman, D. G., Egger, M., Pocock, S. J., Gøtzsche, P. C., & Vandenbroucke, J. P. (2007). The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies. PLoS Medicine, 4(10), e296. https://doi.org/10.1371/journal.pmed.0040296