The influence of <scp>bioRχiv</scp> on <scp> PLOS ONE </scp> 's <scp>peer‐review</scp> and acceptance time

Hiroyuki Tsunoda, Yuan Sun, M. Nishizawa, Xiaomin Liu, Kou Amano

TL;DR

This study reveals that posting a research paper to a preprint server like bioRxiv before submitting it to a journal like PLOS ONE significantly speeds up the peer-review and acceptance process. This suggests that preprints, by allowing for early feedback and initial vetting, can streamline scientific publishing.

What Problem Does This Paper Address?

The traditional academic publishing process, while crucial for ensuring quality, is often criticized for its lengthy delays. This can hinder the rapid dissemination of new scientific findings. The advent of preprint servers, where researchers can share their work before formal peer review, has introduced a new dynamic. This paper investigates how the timing of posting to a preprint server relative to journal submission impacts the subsequent peer-review and acceptance time for a major open-access journal, PLOS ONE. The core question is: does going the preprint route first make the journal publication process faster?

Methods

The researchers employed a multi-step approach:

  1. bioRxiv Data Collection: They used a web crawler to gather metadata for all papers posted on bioRxiv.
  2. PLOS ONE Data Collection: They also collected data on articles published in PLOS ONE.
  3. Data Merging: The crucial step was matching papers from bioRxiv with their corresponding publications in PLOS ONE, primarily using published DOIs.
  4. Defining Peer-Review Time (PT): PT was calculated as the difference between the accepted date and the received date of a paper in the journal.
  5. Categorizing Posting Order: Papers were categorized into three groups based on the timing of preprint posting (PP) and journal submission (JS):
  6. Same-Time: PP and JS were within 7 days of each other.
  7. Preprint-First: PP occurred more than 7 days before JS.
  8. Journal-First: JS occurred more than 7 days before PP.

Datasets

  • bioRxiv: A preprint server for biological sciences. The study analyzed metadata from papers posted on bioRxiv.
  • PLOS ONE: A major open-access scientific journal. The study analyzed published articles from PLOS ONE.

The study matched papers from bioRxiv to PLOS ONE, creating a dataset of articles that had a preprint version before or around their journal submission.

Key Insights and Ideas

  • Preprint First is Faster: The most striking finding is that papers posted to bioRxiv before journal submission to PLOS ONE had a significantly shorter median peer-review and acceptance time.
  • Early Feedback Loop: This suggests that the early feedback and initial vetting that preprints provide can streamline the journal's review process, potentially by addressing minor issues or clarifying the work before formal submission.
  • bioRxiv as a Vetting Tool: The study implies that bioRxiv acts as a preliminary screening mechanism, potentially reducing the burden on journal editors and reviewers by filtering out less robust submissions or allowing authors to refine their work based on early community input.
  • Journal-First is Slower: Conversely, papers submitted to the journal first, without a preceding preprint, experienced longer peer-review and acceptance times.

Key Results

  • Median PT:
    • Preprint-First: 110.00 days
    • Journal-First: 139.50 days
  • Statistical Significance: A Mann-Whitney U test confirmed that the median peer-review and acceptance time for Preprint-First papers was significantly different (and shorter) than for Journal-First papers.
  • Data Distribution: The peer-review and acceptance times for all groups were found to be non-normally distributed, making the median a more suitable measure of central tendency.
  • PLOS ONE's Role: PLOS ONE was identified as a top journal for papers posted on bioRxiv, highlighting its importance in the open-access and preprint ecosystem.

Limitations and Future Work

  • Focus on PLOS ONE: The study's findings are specific to PLOS ONE. The authors suggest further research analyzing multiple journal types (open access and hybrid) to see if these trends hold across the broader publishing landscape.
  • Statistical Data Only: The current study relied solely on statistical data. The authors recommend incorporating qualitative investigations to gain a deeper understanding of the peer-review process and how preprints influence it.
  • Potential for Bias: While not explicitly stated as a limitation, the study's reliance on matching papers might have inherent biases. The authors also note that PLOS ONE's pre-submission screening for plagiarism and nonscientific content might contribute to shorter PTs, a factor that could influence the observed differences.

Summary updated

Abstract

Abstract This study examines the relation between acceptance times in preprint publishing and journal publishing. Specifically, we investigated the association between a paper's posting time to bioRχiv, a preprints server, and journal articles' peer‐review and acceptance time for PLOS ONE . So far, of the total papers published in 1,626 academic journals, the average publication rate of those posted in bioRχiv is 40.67%. Meanwhile, PLOS ONE was the journal that published more papers. Analysis of peer‐review and acceptance time of papers published in journals via preprints showed the time these papers are posted in relation to these intervals. The median of the peer‐review and acceptance time of the journal submission date that was later than the date of first posting to bioRχiv was 110.00 days, and in the reverse case, it was 139.50 days. Posting to the preprint server before journal submission shows a better order than vice versa. This study provides us a good understanding of the peer‐review process. It also gives us good insights into optimizing this process, which would then facilitate paper publication and knowledge dissemination.
BibTeX
@inproceedings{tsunoda2020influence,
  author = {Hiroyuki TSUNODA and Yuan Sun and M. Nishizawa and Xiaomin Liu and Kou Amano},
  title = {{The influence of <scp>bioRχiv</scp> on <scp> <i>PLOS ONE</i> </scp> 's <scp>peer‐review</scp> and acceptance time}},
  booktitle = {Proceedings of the Association for Information Science and Technology},
  year = {2020},
  doi = {10.1002/pra2.398},
}

Loading related content...

Figure