What Is a Peer-Reviewed Source, and How to Find One

What “peer-reviewed” actually means

A peer-reviewed article is one that went through a formal check by experts before it was published. The process usually runs like this. Researchers submit a manuscript to a journal. An editor screens it for fit and basic quality, and turns many submissions away at this stage. If it passes, the editor sends it to two or more reviewers who work in the same field but were not involved in the study. They read it closely and write reports on its methods, evidence, reasoning and contribution.

Based on those reports, the editor decides whether to accept the article, reject it, or send it back for minor or major revisions. Revised manuscripts often go through another round. Only then is the article published, often with dates showing when it was received, revised and accepted. You will see the same idea described as “refereed” or, more loosely, “scholarly,” though not every scholarly publication is refereed.

Which sources are peer reviewed, and which aren't

The phrase “peer-reviewed journal” causes confusion because journals publish more than research articles. A book review or editorial in a respected journal is not a peer-reviewed source, even though the journal is. Use this table as a guide, and check the specific item when it matters.

When an assignment says “peer-reviewed sources only,” it almost always means research and review articles from refereed journals. If you want to use a scholarly book or a conference paper, ask your instructor first.

How to find peer-reviewed sources, step by step

The quickest route to peer-reviewed sources is your library's databases, which index scholarly journals and often let you filter by review status. Your library's website lists the databases you can use, usually grouped by subject.

Strong searches make step two far more productive; the guide to Boolean search operators shows how to combine terms so the filter has good results to work on.

1. 1Pick a database that fits your subject: PubMed for medicine and health, APA PsycInfo for psychology, ERIC for education, JSTOR for the humanities and social sciences, or a multidisciplinary database your library subscribes to.
2. 2Search with a few key terms joined by AND, add synonyms with OR, and use quotation marks to keep phrases together.
3. 3Turn on the filter labeled “Peer reviewed,” “Scholarly (peer reviewed) journals” or similar. Not every database has one; where it is missing, check each journal yourself.
4. 4Limit the document type to articles, which removes most book reviews, editorials and news items.
5. 5Read the abstracts of promising results to confirm each article actually addresses your question.
6. 6For anything you plan to rely on, confirm the journal's review policy using the checks below.

Finding peer-reviewed sources with Google Scholar

Google Scholar is a convenient place to search, but it has no peer-review filter. Its results mix journal articles with preprints, theses, books, conference papers, course materials and citations to works it cannot show you. A result appearing in Scholar tells you it looks academic, not that experts reviewed it.

Use Scholar for discovery, then check each result as you would anywhere else: identify the journal and confirm its policy. Two features help. “Cited by” lists later work that cites an article, which is useful for finding newer research. And if you add your university under Library links in Scholar's settings, results your library subscribes to show a link to the full text. The guide on using Google Scholar effectively covers both in more detail.

How to tell if an article is peer reviewed

Database filters are useful but imperfect, so check any source you will lean on heavily. Start with the most reliable test and work down:

If you still cannot tell, ask a librarian. Confirming whether a journal is refereed is a routine question for them, and they can usually answer it in minutes.

• The journal's website: look under About, Aims and Scope, Editorial Policies or Instructions for Authors for a description of the review process. This is the firmest evidence.
• Ulrichsweb, a periodicals directory many libraries subscribe to, marks refereed journals.
• The article's history: received, revised and accepted dates usually indicate a review process.
• The article type: a label such as “Research article” or “Original investigation” rather than “Editorial,” “Commentary” or “Letter.”
• The publisher: a university press, scholarly society or established academic publisher is a good sign, though not proof on its own.
• The form: an abstract, a methods section, close argument and a full reference list are typical of reviewed work, but predatory journals copy this look.

Watch out for predatory journals

Predatory journals charge authors to publish while providing little or no real review, and their articles can look convincing in a search result. They matter to you because an article from one can appear in Google Scholar right beside legitimate research. Warning signs include:

For open-access journals, one useful check is whether the journal is listed in the Directory of Open Access Journals (DOAJ), which screens journals before including them. Absence from that directory is not proof of a problem, but presence is reassuring.

• Promises of unusually fast review or acceptance, sometimes within days.
• Mass email invitations to submit, often flattering and only loosely related to the recipient's field.
• A very broad scope, such as one journal covering engineering, medicine and the arts.
• A name that closely imitates a well-known journal.
• An editorial board you cannot verify.
• Unclear fees, or a website with no clear account of how review works.

Peer review is a filter, not a guarantee

Peer review raises the odds that an article is sound, which is why instructors ask for it. But reviewers do not usually rerun experiments or reanalyze raw data; they judge what the manuscript shows them. Honest mistakes, weak statistics and occasionally fraud can get through. Journals publish corrections, and seriously flawed articles can be retracted, so check that an article you rely on has not been; the article page or database record will usually say so.

Reviewed findings can also fail to hold up when other researchers try to repeat them, especially when they rest on a single study. Treat peer review as the start of evaluation, not the end of it: ask how large the study was, whether other work agrees, and whether the conclusion matches the evidence. The guide to evaluating a source gives a fuller checklist.