Direct comparison
Preprints vs. Peer-Reviewed Articles
Preprints are unreviewed manuscripts; peer-reviewed articles are formally vetted. Compare stability, citability, and how funders/tenure treat each.
Written and maintained by CASRAI Editorial Board
Last updated
Ask CASRAI · free to try
Ask about Preprints vs. Peer-Reviewed Articles
Ask your first 2 questions free below. Subscribers get 150 a day for $29 a month.
An AI assistant specialized in research administration. It cites the sources behind every answer, labels web answers and says when it can't answer.
Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.
Works on this site and inside Claude, Cursor and the AI tools you already use.
Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays free to read, with no account and no card.
How do Preprint, Peer-Reviewed Article compare side by side?
The table below compares Preprint, Peer-Reviewed Article across 11 procurement-relevant dimensions, from what it is through how to cite it.
Side-by-side comparison
| Dimension | Preprint | Peer-Reviewed Article |
|---|---|---|
| What it is | The author's complete manuscript, posted publicly before or without journal peer review | The manuscript after independent expert review, revision, and formal editorial acceptance |
| Peer review status | Not independently peer reviewed (most servers do a screening check for scope/plagiarism/ethics, not scientific validity) | Reviewed by independent subject-matter experts selected by a journal editor, under a defined review model |
| Typical time to public availability | Hours to a few days after server screening | Weeks to many months, depending on field and journal (review rounds, revisions, production) |
| Version stability | Can be revised by the author; most servers version it explicitly (v1, v2, v3...) and keep all versions accessible | Fixed once published as the version of record; further changes require a formal correction, erratum, or retraction notice |
| Persistent identifier | Usually has its own DOI, issued by the preprint server, distinct from any later journal DOI | Has a DOI issued by the journal/publisher via Crossref, treated as the canonical citable identifier |
| Where it's hosted | A preprint server (arXiv, bioRxiv, medRxiv, SSRN, ChemRxiv, Research Square, etc.) or an open-science platform (OSF) | The publisher's own platform, and often mirrored in PubMed Central, Europe PMC, or an institutional repository |
| Cost to make public | Free to post on nearly all major servers | Free to read only if gold/diamond/hybrid open access or self-archived; otherwise behind a subscription paywall unless the author or their institution pays an APC |
| Treatment in grant applications | NIH explicitly permits citing preprints and other interim research products in applications and progress reports (NOT-OD-17-050, effective May 2017); many other funders now allow this too | Universally acceptable as a cited research product across all funders |
| Treatment in tenure and promotion review | Not yet standard evidence of research quality at most institutions; increasingly accepted as a marker of productivity and priority, especially under DORA/CoARA-aligned narrative-CV practices | The default currency of tenure and promotion review, though DORA and CoARA both push committees to weigh the work itself over the journal's prestige |
| Retraction, correction, or withdrawal | Can be withdrawn by the author at any point; servers post a withdrawal notice but do not conduct a formal editorial retraction investigation | Corrections and retractions follow COPE-aligned editorial process and are permanently linked to the version of record |
| How to cite it | Cite as a preprint, name the server, and include the preprint DOI -- always check first whether a peer-reviewed version now exists and cite that instead if it does | Cite the journal, volume/issue, page or article number, and the article's own DOI |
Common questions
Common questions about Preprint vs Peer-Reviewed Article
Can I cite a preprint in a manuscript or grant application?
+
Yes, in most cases. Journals increasingly permit citing preprints (clearly labeled as such), and funders including NIH explicitly allow citing preprints and other interim research products in applications and progress reports. Always check the specific journal's or funder's current citation policy, and cite the peer-reviewed version instead if one has since been published.
Do preprints count toward tenure or promotion?
+
It depends on the institution. Preprints are not yet standard, weighted evidence in most tenure dossiers, but committees operating under DORA or CoARA-aligned assessment practices are more likely to consider a preprint as evidence of productivity, priority, or open-research practice, particularly when a narrative CV format is used.
Is a preprint peer reviewed at all?
+
No, not in the formal sense. Most preprint servers run an internal screening check for topical scope, plagiarism, and basic ethical/legal issues before posting, but this is not independent expert review of the science, and it is explicitly not a substitute for journal peer review.
What happens if a preprint is later rejected by a journal?
+
The preprint typically stays publicly available on the server unless the author withdraws it. A journal rejection does not automatically trigger removal, and most servers do not require authors to disclose rejection. This is one reason preprints should never be treated as equivalent in reliability to a peer-reviewed article.
How can I tell whether a specific article I'm reading is a preprint or the peer-reviewed version?
+
Preprint servers label content clearly (a banner noting it has not been peer reviewed, plus a version number). Peer-reviewed articles carry a journal name, volume/issue, and a publisher-issued DOI. When in doubt, search the title in the journal's own site or Crossref to check whether a peer-reviewed version of record now exists.
Going deeper
