How to Know If an Article Is Peer Reviewed

“Peer reviewed” is not a label printed on an article — it’s a claim about the process an article went through before publication, and verifying that claim takes a few concrete checks rather than a guess based on how polished a PDF looks. This guide walks through the practical methods research administrators, librarians, authors, and reviewers actually use to confirm peer-review status, how to avoid being fooled by venues that only look peer reviewed, and why “peer reviewed” itself isn’t a single, binary thing.

The quick answer

No single check is fully sufficient on its own — combine at least two:

1. Read the journal’s own editorial policy or “About” page. A genuine peer-reviewed journal states its review process explicitly.
2. Use a library database’s peer-reviewed/refereed filter, or check Ulrichsweb’s “Refereed” indicator for the journal title.
3. Check whether the journal is indexed in a database that requires peer review as a condition of indexing, such as MEDLINE — treating this as a strong signal, not an automatic guarantee.
4. Confirm it isn’t a predatory or pay-to-publish venue exploiting the credibility peer review normally signals.

1. Check the journal’s own editorial policy or “About” page

Every legitimate peer-reviewed journal publishes a description of its review process, usually under a page labeled “About,” “Editorial Policies,” “Peer Review Process,” or similar in the author/reviewer guidelines. A credible policy page will typically specify:

• Whether submissions are sent to external subject-matter experts (not just handled internally by an editor)
• How many independent reviewers are typically assigned per manuscript
• The review model used — single-anonymized, double-anonymized, or open (see the section below)
• Named editors or an editorial board with verifiable institutional affiliations

A vague statement (“all submissions are carefully reviewed”) without specifics on reviewer independence or process is a weaker signal than a page that names a concrete workflow. This check is a necessary starting point, but it’s self-reported by the journal, which is exactly why it should be combined with an independent, third-party check.

2. Use a library database’s peer-reviewed filter, or check Ulrichsweb

Most academic library databases — including EBSCO/ProQuest platforms, Web of Science, Scopus, and Google Scholar’s advanced search — let you restrict results to peer-reviewed (sometimes labeled “refereed”) journals directly in the search interface. This is usually the fastest single check available to a researcher already searching a database.

For checking a specific journal title rather than filtering a search, Ulrichsweb (Ulrich’s Periodicals Directory) is the standard library reference tool. It displays a distinct “Refereed” icon — a referee’s-jersey graphic — next to a journal’s record once its peer-review status has been independently confirmed. Ulrich’s content team verifies this by contacting publishers directly, reviewing the journal’s own editorial policies, and corroborating with librarians and editorial board members; the icon is deliberately withheld until that verification is complete, so its presence is a meaningfully vetted signal rather than a self-reported one. Most academic libraries provide institutional access to Ulrichsweb — check with a research or subject librarian if you don’t have a direct login.

3. Check indexing in a peer-reviewed-only database — and know its limits

Being indexed in a database that requires peer review as a condition of inclusion is a strong, independently-verified signal — but it is a proxy, not a certification of any individual article’s review quality, and it’s worth understanding exactly what the proxy does and doesn’t confirm.

MEDLINE (the U.S. National Library of Medicine’s biomedical literature database, searchable via PubMed) is the clearest example in the health and life sciences. NLM’s journal selection process is a “rigorous, multi-step process” that evaluates a journal comprehensively rather than against a single checklist — but a clearly stated, sufficiently detailed peer-review policy (including the type of review used and the typical number of reviewers per manuscript) is explicitly one of the criteria assessed, alongside editorial quality, scientific merit, and production standards. Historically this evaluation was informed by a standing NIH-chartered advisory body, the Literature Selection Technical Review Committee (LSTRC); NLM disbanded LSTRC in September 2025 as part of a broader modernization of its journal-review process, replacing the committee structure with external subject-matter and library-science consultants. The underlying principle is unchanged: MEDLINE indexing still functions as a meaningful, externally-vetted quality signal, but it reflects NLM’s assessment of the journal as a whole, not a guarantee about the rigor of review applied to any specific article in it.

The same logic extends to other selective, peer-reviewed-only indexes relevant outside biomedicine — Scopus and Web of Science both apply editorial and peer-review criteria before accepting a journal for coverage, and the Directory of Open Access Journals (DOAJ) requires a documented peer-review process as a condition of listing open-access titles. Indexing in any of these is a reasonable proxy precisely because each index runs its own vetting process before adding a title — but none of them re-verify that every single published article actually went through rigorous review, which is why this check works best combined with the other methods on this page rather than used alone.

4. Distinguish peer review from editorial review

These two terms are often used loosely and interchangeably, but they describe different processes, and the distinction matters when you’re trying to verify a specific claim:

• Peer review means the manuscript was evaluated by independent experts in the same field (“peers”) who were not involved in producing the work — reviewers assess the methodology, evidence, and conclusions, and typically recommend accept, revise, or reject.
• Editorial review (sometimes called “internal review”) means an in-house editor, or a small editorial team, made the accept/reject decision without sending the manuscript to external subject-matter reviewers. Editorial review can still be careful and professional — many magazines, trade publications, and some journal front-matter (editorials, commentaries, letters) go through editorial review only — but it is not peer review, and describing it as such is inaccurate.

This is exactly the ambiguity ANSI/NISO Z39.106-2023, Standard Terminology for Peer Review, was developed to resolve. Published by NISO in July 2023 (based on a taxonomy originally developed by the STM Association’s peer-review working group), the standard defines a controlled vocabulary across four dimensions — identity transparency between author and reviewer, who the reviewer interacts with, what review information is published, and whether post-publication commenting is enabled — specifically so that “peer review,” “editorial review,” and specific review models aren’t used as loosely interchangeable marketing terms. See CASRAI’s ANSI/NISO Z39.106-2023 dictionary entry for the full terminology breakdown.

A related, often-overlooked distinction: not every item in a peer-reviewed journal is itself peer reviewed. Editorials, letters to the editor, corrections, book reviews, and some commentaries are commonly published without going through the same external review as original research articles, even in a genuinely peer-reviewed journal. Checking an individual article’s type — not just the journal’s overall status — matters for this reason.

5. Remember that “peer reviewed” isn’t binary

A common misconception is that peer review is one uniform process, so an article either “has it” or doesn’t. In practice, peer review comes in several distinct models, and all of them count as genuine peer review — they differ in what information is disclosed between authors and reviewers, not in whether independent expert evaluation happened:

• Single-anonymized review (formerly called “single-blind”): reviewer identities are hidden from authors, but reviewers know who the authors are.
• Double-anonymized review (formerly called “double-blind”): both author and reviewer identities are withheld from each other, at least in principle.
• Open review: reviewer identities, review reports, or both are disclosed — sometimes published alongside the final article.

None of these models is more or less “real” peer review than the others — they’re different transparency trade-offs the journal or publisher has chosen, and a journal’s use of open review shouldn’t be mistaken for a lack of rigor, nor should single-anonymized review be assumed to be the only legitimate default. For a full side-by-side comparison of how these models differ in practice, rationale, and trade-offs, see CASRAI’s single-blind vs. double-blind vs. open peer review comparison.

6. Watch for predatory or pay-to-publish venues faking peer review

A venue can claim to be peer reviewed, use the language of peer review in its marketing, and still not actually provide the service — this is the defining feature of a predatory journal: a publication that solicits and accepts manuscripts in exchange for publication fees without providing the editorial and peer-review services normally expected of a legitimate scholarly venue, often through deceptive marketing about its indexing, impact factor, or editorial board.

Practical red flags worth checking specifically because they undermine a predatory venue’s claim to genuine peer review:

• Unusually fast turnaround (acceptance within days, regardless of manuscript quality)
• An editorial board that can’t be independently verified, or that lists researchers who have never agreed to serve
• Aggressive, unsolicited email solicitation to submit or to guest-edit
• No clear statement of the review process, or a description too vague to evaluate
• Publication fees not disclosed until after acceptance

Cross-checking against the methods above is the most reliable defense: a real peer-reviewed journal should show up with a legitimate refereed status in Ulrichsweb, have a plausible indexing record, and describe a specific, verifiable review process — not just assert one. For a full walkthrough of identifying predatory venues, see CASRAI’s guide, How to Identify Predatory Journals and Publishers, and the Think.Check.Submit checklist maintained by a coalition of scholarly-publishing organizations.

A step-by-step verification checklist

1. Look up the journal’s editorial policy or “About” page and check whether it names a specific external review process.
2. Search the journal title in Ulrichsweb (via your library) and check for the “Refereed” indicator.
3. Check whether the journal is indexed in a relevant peer-reviewed-only database — MEDLINE/PubMed for biomedicine, or Scopus/Web of Science/DOAJ more broadly.
4. Confirm the specific article type — original research and reviews are typically peer reviewed; editorials, letters, and commentaries often are not, even in the same journal.
5. Check the review model disclosed (single-anonymized, double-anonymized, or open) — any of the three is legitimate peer review.
6. Screen for predatory-journal red flags if anything about the venue feels unfamiliar or the invitation to submit was unsolicited.

Frequently asked questions

Does Google Scholar tell me if an article is peer reviewed?

Not directly. Google Scholar indexes a very broad range of scholarly and non-scholarly material and doesn’t apply a peer-review filter of its own the way a curated academic database does. Its “advanced search” doesn’t include a dedicated peer-reviewed filter either — for that, use a library database (EBSCO, ProQuest, Web of Science, Scopus) or check the journal directly via Ulrichsweb.

Is a preprint peer reviewed?

No. A preprint is a manuscript posted to a server such as arXiv, bioRxiv, or medRxiv before, or independent of, formal journal peer review. Preprints are explicitly not peer reviewed at the point of posting — that’s the defining feature of the format — even though many preprints are later peer reviewed and published in a journal.

If a journal has an impact factor, does that mean it’s peer reviewed?

Not necessarily, though it’s a reasonably strong indirect signal. Having an impact factor means the journal is indexed in a citation database that calculates one (typically Web of Science, for a Journal Impact Factor), and those indexes apply their own editorial vetting before including a title — but an impact factor is a citation metric, not a peer-review certification, and it says nothing about the review applied to any individual article.

Can conference proceedings be peer reviewed?

Yes. Many academic conferences, particularly in computer science and engineering, use rigorous peer review for accepted papers — sometimes more selective than journal review. Peer review status for conference proceedings should be verified the same way: check the conference’s own submission/review policy rather than assuming based on venue prestige alone.

How is peer review different from editorial review?

Peer review sends a manuscript to independent external experts in the same field for evaluation. Editorial review is a decision made in-house by an editor or editorial team without external subject-matter reviewers. Both can be legitimate processes, but only the first is “peer review” in the accurate sense — see the distinction above.

For the broader mechanics of how peer review works end to end, see CASRAI’s Peer Review: What It Is, How It Works, and Why It Matters guide, and for the reviewer’s side of the process, Peer Reviewer: Role, Responsibilities, and How to Become One.