The abstract is the most-read section of any research paper. It determines whether a reader downloads the full text or moves on. Despite being only 150–250 words, it is often the hardest part to write well.
What an Abstract Must Cover
Think of the abstract as a miniature version of your paper. It should answer five questions in order:
- Background — Why does this topic matter? (1–2 sentences)
- Objective — What did you set out to investigate? (1 sentence)
- Method — How did you approach it? (1–2 sentences)
- Results — What did you find? (2–3 sentences)
- Conclusion — What does it mean? (1 sentence)
The Formula
Use this template as a starting point:
[Topic] is important because [reason]. This study investigates [research question] using [method/data]. The results show that [key finding 1] and [key finding 2]. These findings suggest that [implication], which has implications for [field/practice].
Adjust the sentence count based on your discipline's norms and word-limit requirements.
Common Mistakes
- Too vague — "This paper studies climate change." What aspect? What method? What finding?
- Including citations — Abstracts should stand alone; avoid references unless the journal allows it.
- Using jargon — The abstract is often read by people outside your exact sub-field. Keep it accessible.
- Copying the introduction — The abstract summarises the entire paper, not just the opening.
- Missing results — Stating you "analysed" something without revealing findings is incomplete.
Example Abstract
Diabetic retinopathy (DR) is a leading cause of preventable blindness, yet screening coverage in rural India remains below 20%. This study evaluates the accuracy of a lightweight convolutional neural network (MobileNetV3) trained on 12,000 fundus images from three public hospitals. The model achieved a sensitivity of 94.2% and specificity of 91.7% on a held-out test set, outperforming traditional image-processing baselines by 8 percentage points. Grad-CAM visualisations confirmed that the network focuses on clinically relevant lesion areas. These results demonstrate that affordable, smartphone-based DR screening is feasible and could significantly expand access in resource-limited settings.
When to Write It
Write the abstract last — after the rest of the paper is complete. This way, you know exactly what the paper contains and can summarise it faithfully.