I analyzed a random sample of 61,458 full-text research papers, uploaded to PubMed Central between the years 2016 and 2021, in order to answer the questions:
What is the typical length of a results section? and which factors influence it?
I used the BioC API to download the data (see the References section below).
Here’s a summary of the key findings
1. The median results section was 992 words long (equivalent to 37 sentences, or 7 paragraphs), and 90% of the results sections were between 285 and 3,258 words.
2. Compared to other sections in a research paper, the results was about the same length as either the methods or the discussion, and double the length of the introduction.
3. Review articles have longer results sections compared to original research articles.
4. In general, articles published in higher impact journals include a more detailed description of their results.
Overall length of the results section
Here’s a table that describes the length of a results section in terms of words, sentences, and paragraphs:
| Results Section Length | |||
|---|---|---|---|
| Word Count | Sentence Count | Paragraph Count | |
| Minimum | 15 words | 2 sentence | 1 paragraph |
| 25th Percentile | 609 words | 22 sentences | 5 paragraphs |
| 50th Percentile (Median) | 992 words | 37 sentences | 7 paragraphs |
| Mean | 1,285.8 words | 50.8 sentences | 9.6 paragraphs |
| 75th Percentile | 1,662 words | 65 sentences | 11 paragraphs |
| Maximum | 38,146 words | 3,336 sentences | 1,714 paragraphs |
From these data, we can conclude that the results sections in most research papers are between 609 and 1,662 words long (22 to 65 sentences).
If you are interested, here are the links to the articles with the shortest and longest results sections.
The results section constitutes 26.2% of the total word count in a research article, equivalent to the length of either the methods or the discussion, and double the length of the introduction [source: How Long Should a Research Paper Be?].
Length of the results for different article types
The following table shows the median word count of the results section for different study designs:
| Study design | Number of studies in the sample | Median results word count |
|---|---|---|
| Case-control | 443 studies | 657 words |
| Case series | 140 studies | 679 words |
| Cross-sectional | 3,528 studies | 683 words |
| Randomized controlled trial | 841 studies | 718 words |
| Pilot study | 686 studies | 750 words |
| Quasi-experiment | 144 studies | 751 words |
| Cohort | 5,178 studies | 855 words |
| Meta-analysis | 1,480 studies | 1,032 words |
| Case report | 400 studies | 1,054 words |
| Systematic review | 689 studies | 1,168 words |
The data show that, in general, original research articles have shorter results sections compared to review articles (i.e. systematic reviews and meta-analyses).
Length of the results in different journals
In order to study the influence of the journal quality on the length of the results section, I ran a Poisson regression that models the results word count given the journal impact factor. Here’s the model output:
| Variables | Coefficient | Standard error | p-value |
|---|---|---|---|
| (Intercept) | 6.906 | <0.001 | <0.001 |
| Journal impact factor | 0.077 | <0.001 | <0.001 |
The model shows that a higher journal impact factor is associated with a longer results section. Specifically, a 1 unit increase in the journal impact factor is associated with an increase of approximately 8% in the results word count. For the median article, this means that a 1 unit increase in the journal impact factor is associated with an increase in 79 words in the results section.
This effect may be explained by the fact that review articles (that have longer results sections) get published in higher-quality journals than other types of articles. But even after controlling for the study design, the coefficient of “Journal impact factor” was 0.076 (very close to the raw effect without adjustment).
So the data suggest that, in general, articles published in higher impact journals include a more detailed description of their results.
References
- Comeau DC, Wei CH, Islamaj Doğan R, and Lu Z. PMC text mining subset in BioC: about 3 million full text articles and growing, Bioinformatics, btz070, 2019.
Further reading
- How Long Should the Introduction of a Research Paper Be?
- How Long Should the Methods Section Be? Data from 61,514 Examples
- How Long Should a Discussion Section Be? Data from 61,517 Examples
- How Many References to Cite? Based on 96,685 Research Papers
- How Old Should References Be? Based on 3,823,919 Examples
- Length of a Conclusion Section: Analysis of 47,810 Examples