Preprints impact factor is a myth: bioRxiv, medRxiv, and other preprint servers cannot carry a Clarivate Journal Impact Factor, because that metric is calculated only for peer-reviewed journals indexed in Web of Science. Preprints are unreviewed manuscripts, not journals, so the correct substitutes are altmetrics, usage data, or — once published — the Impact Factor of the destination journal.
A preprint is a complete draft of a research manuscript that is made publicly available, typically via a dedicated server such as bioRxiv, medRxiv, or SSRN, before it has undergone formal peer review.
- What Is the “Preprints Impact Factor” Search Actually Asking?
- Why bioRxiv and medRxiv Cannot Carry a Clarivate Impact Factor
- The bioRxiv Scimago (SJR) Confusion, Explained
- Legitimate Metrics to Cite Instead of an Impact Factor
- Frequently Asked Questions
- What This Means for Authors, Institutions, and Publishers
What Is the “Preprints Impact Factor” Search Actually Asking?
Most searchers typing “biorxiv impact factor” or “medrxiv impact factor” are trying to answer one practical question: is this a legitimate place to read or deposit research, and how citable is it? They are borrowing a journal-quality shorthand and applying it to infrastructure that was never designed to carry one.
bioRxiv’s own FAQ answers the question directly and definitively: “bioRxiv is not a journal so it has no Impact Factor.” The same logic applies to medRxiv, SSRN, Preprints.org’s raw submissions, and every other preprint repository — none of them are journals in the bibliometric sense, regardless of how large, credible, or heavily used they are.
Why bioRxiv and medRxiv Cannot Carry a Clarivate Impact Factor
The Journal Impact Factor is a proprietary metric published annually in Clarivate’s Journal Citation Reports (JCR). It divides the citations a journal received in a given year by the number of citable items (articles and reviews) that journal published in the two preceding years. Calculating it requires three things a preprint server structurally lacks:
- Editorial peer review — JCR-indexed titles must operate a formal review process; bioRxiv and medRxiv post manuscripts after only a basic screening for plagiarism, ethics, and non-scientific content, not scientific validation.
- A fixed, versioned “journal” corpus — Impact Factor counts citations to a stable set of articles published under one journal title. Preprints are living documents that are frequently revised, withdrawn, or superseded by a published version with a different title and author list.
- Web of Science indexing as a source title — Clarivate assigns an Impact Factor only to titles it has selected as journal source publications for JCR. Preprint servers are indexed as repositories, not journal source titles, so they are structurally ineligible.
bioRxiv was co-founded by John Inglis and Richard Sever in November 2013 and is now operated by openRxiv, a non-profit dedicated to advancing science communication. medRxiv followed in June 2019, co-sponsored by Cold Spring Harbor Laboratory, Yale University, and BMJ. Both were built explicitly as pre-peer-review distribution services, not as journals competing for a JCR listing.
As Elmore et al. (2018) put it in a widely cited PMC analysis: “Because they are not journals, preprint servers have no Impact Factor and authors retain copyright of their articles.” That is a structural fact, not a temporary gap that a future JCR update will fill.
The bioRxiv Scimago (SJR) Confusion, Explained
The same logic rules out the SCImago Journal Rank (SJR) and Elsevier’s CiteScore — searchers sometimes look for “biorxiv scimago” expecting an equivalent, Scopus-based alternative to the Clarivate figure. It does not exist either, for an identical structural reason.
SJR and CiteScore are both derived from Scopus, and Scopus’s source-inclusion criteria select serial publications with editorial and peer-review policies — the same gate that excludes bioRxiv and medRxiv from JCR. Preprint repositories are not evaluated as candidate “sources” in Scopus at all, so no SJR or CiteScore figure is ever generated for them.
This is distinct from Preprints.org, an MDPI-run platform that recommends “friendly journals” carrying real Impact Factors on its own site — a detail that fuels confusion in search results, because the Impact Factor shown belongs to a destination journal, not to the preprint submission itself.
Legitimate Metrics to Cite Instead of an Impact Factor
Preprints are not impact-free, they are simply measured differently. The following metrics are appropriate, real, and already in use across the scholarly-communication ecosystem:
| Metric | Provider | What it measures | Applies to preprints? |
|---|---|---|---|
| Journal Impact Factor | Clarivate (JCR) | Mean citations per citable item, peer-reviewed journals only | No |
| SCImago Journal Rank (SJR) | Scopus / SCImago | Prestige-weighted citations, Scopus-indexed serials only | No |
| Altmetric Attention Score | Altmetric | News, policy, social, and blog mentions of any DOI | Yes |
| Downloads and abstract views | bioRxiv / medRxiv usage stats | Raw engagement with a specific preprint | Yes |
| Eventual journal citation count | CrossRef / Google Scholar / Semantic Scholar | Citations once the manuscript is formally published | Indirectly |
bioRxiv and medRxiv preprints are registered with Crossref DOIs, which makes them citable and part of the permanent scholarly record even without peer review. Academic norms increasingly treat a preprint’s own download count and Altmetric score, plus the citation trajectory of its eventual journal version, as the legitimate proxy for “impact” — never a fabricated Impact Factor for the preprint itself.
Since June 2020, PubMed Central’s NIH Preprint Pilot has indexed preprints resulting from NIH-funded research, clearly labelling them as unreviewed while still making them discoverable — an example of measured, honest visibility rather than a borrowed journal metric.
Frequently Asked Questions
Does bioRxiv count as published?
No. Posting on bioRxiv or medRxiv is not equivalent to formal publication. It establishes a timestamped, DOI-registered public record and priority date, but the manuscript has not passed peer review until it appears in a journal that has independently assessed it.
Is bioRxiv considered a journal?
No. bioRxiv describes itself as an archive and distribution service for unpublished preprints in the life sciences, operated by the non-profit openRxiv. It performs basic screening, not editorial peer review, which is why it cannot be classified as a journal for indexing purposes.
Is it okay to cite bioRxiv?
Yes. Manuscripts posted on bioRxiv receive DOIs and are part of the citable scientific record. Citing organisations, including ICMJE, permit preprint citation provided the source is clearly labelled as not peer-reviewed, so readers can weigh the evidence accordingly.
Are preprints indexed in PubMed?
Selectively. Since 2020, PubMed Central has indexed preprints arising from NIH-funded research under the NIH Preprint Pilot. Where both a preprint and its later journal version exist, PMC links the two records so readers can identify which has undergone peer review.
What This Means for Authors, Institutions, and Publishers
For authors, the practical takeaway is definitional discipline: a preprint’s value is measured by downloads, Altmetric attention, and its eventual journal citation record — not by a number that will never be assigned to the platform itself. Institutions running tenure and promotion review should train evaluators to distinguish these signals explicitly, rather than let a search-engine autocomplete suggestion imply that one exists.
Publishers have already begun formalising alternatives rather than chasing a journal-shaped metric for preprints. eLife’s Reviewed Preprints model, introduced in January 2023, replaces the binary accept/reject decision with public peer reviews and a structured eLife Assessment using a controlled vocabulary of terms — a deliberate move away from Impact-Factor logic rather than an attempt to imitate it.
As preprint volume continues to grow across biology and medicine, expect more publishers and funders to formalise usage-based and altmetric-based reporting standards specifically for unreviewed manuscripts, rather than waiting for Clarivate or Scopus to extend journal-only metrics to infrastructure that was deliberately built to sit outside the peer-review gate. Research administration teams evaluating researcher outputs should build these distinctions into their own research administration policies now, and consult a maintained scholarly-communication dictionary when standardising terminology across review committees.
Leave a Reply