A team had one tidy result: a widely used scale behaved very differently in first-year students than in final-year students. They tried to turn it into a 7,000-word paper, padding it with two secondary questions to make it look substantial. Reviewers said the paper felt stretched and its strongest point was buried under weaker analyses. As a 2,500-word brief report it would very likely have moved much faster.
Most journals accept several article types, each with its own length limits, structure and expectations. Choosing the right one is a strategic decision, not a box to tick in the submission system.
A map of common article types
| Type | Typical length | Best for |
|---|---|---|
| Original research article | Longest, within the journal’s limit | A complete study with full question, design, results and discussion |
| Short communication or brief report | About a third to half of a full paper | One clear, contained finding needing little context |
| Research note | Very short | Small results, null findings, preliminary observations, replications |
| Review | Usually long | Systematic or narrative synthesis; many journals publish reviews by invitation only |
| Methods paper | Medium | A new tool, procedure, software or measure with evaluation |
| Data paper or data descriptor | Short to medium | Describing a published dataset so others can reuse it |
| Commentary or perspective | Short | An argument about a field-level issue; often invited |
| Letter to the editor | Very short | A response to an article published in that journal |
| Registered report | Like a full paper, submitted in two stages | Studies reviewed and accepted in principle before data collection |
Names and limits vary between journals. Always read the “Article types” section of the author guidelines for the specific journal you are targeting.
Short reports: when less is stronger
A short report fits when the finding can be stated in one sentence and supported by one or two tables. The advantages: reviewers read it quickly, turnaround is often faster and the finding is not diluted. The cost: you lose most of the context, the introduction and discussion shrink to a few paragraphs, and reference limits are often tight.
Do not use a short report to rush out a study that needs space. Reviewers will notice methods so compressed they cannot be checked.
Data papers: credit for collecting the data
Collecting and cleaning a good dataset can take years, yet in a research article it earns one methods paragraph. A data paper gives that work its own citation. It typically covers:
- What was measured, where, when and on whom.
- Collection procedures and quality control.
- File structure, codebook and formats.
- Repository, persistent identifier and licence.
- Suggested reuse and known limitations.
Data papers usually do not test hypotheses. If you want to show findings, that is a different paper.
Registered reports: peer review before the results
In this model you submit the introduction, methods and analysis plan before collecting data. If the journal grants in-principle acceptance at stage one, it commits to publishing the results whether or not they support the hypotheses, provided you follow the plan. It suits studies where a null result would still be informative, and researchers who want design feedback before investing in data collection. The drawbacks: the first review round takes time, and it does not suit exploratory work without clear hypotheses.
Null results and replications: out of the drawer
Many “no effect” results are never written up because authors assume no journal will take them. In fact research notes, short reports and registered reports are all suitable routes. A null result from an adequately powered, well-designed study has real value: it stops others walking down the same dead end.
Commentaries and letters: use them at the right moment
A commentary is not a home for a small study that lacks data. It is a place to argue, usually from experience or synthesis. Early-career researchers can write one if they have a sharp point and good evidence, but check whether the journal accepts unsolicited commentaries. Letters to the editor are time-bound: many journals accept them only within weeks or months of the original article.
In practice: five questions to choose a type
- Can my main finding be stated in one sentence and shown in one or two tables? If so, consider a short report.
- Is the main value in the data, the tool or the finding? Data suggests a data paper; a tool suggests a methods paper.
- Have I collected the data yet? If not, and I have clear hypotheses, consider a registered report.
- Is the result null or a replication? Look for journals with article types that welcome them.
- Am I arguing or reporting? Arguing is a commentary; reporting is research.
Next step: open the author guidelines of two journals you are targeting and copy their article types and limits onto one page. Put your current draft next to it. Often the answer is obvious without changing a single result.
Câu hỏi thường gặp
What is the difference between a short communication and a full research article?
A short communication presents one contained finding with less context, fewer tables and tighter reference limits. It suits clear results that do not need layered analysis.
What is a data paper?
A data paper describes a published dataset: what was measured, how it was collected, how it is structured and where it is stored. It gives data collectors separate credit and citations.
How does a registered report work?
You submit the introduction, methods and analysis plan before collecting data. If accepted in principle, the journal commits to publishing the results as long as you follow the plan.
Can I publish null or negative results?
Yes. Research notes, short reports and registered reports are all suitable routes, particularly when the study was adequately powered and well designed.
Can early-career researchers write commentaries?
Yes, if they have a sharp argument and good evidence, but many journals publish commentaries only by invitation. Check the author guidelines before writing.
