How to check if a paper has been retracted

You found the perfect paper two years ago. It supported a key claim in your literature review, you cited it in three chapters, and it has been sitting quietly in your reference list ever since. What you may not know is that the journal withdrew it months ago, and nothing in your workflow was ever going to tell you.

Retracted papers do not announce themselves. The PDF on your hard drive looks exactly as it did on the day you downloaded it. Your reference manager stores the metadata it fetched at import and never looks back. Unless you actively check, a retraction can sit undetected in your bibliography all the way to submission, and examiners do notice.

This guide covers what retraction actually means, why retracted work keeps getting cited, how to check a single paper in under a minute, and how to sweep an entire thesis bibliography before you submit.

What a retraction actually means

A retraction is a formal statement by a journal that a published paper should no longer be relied upon. The paper usually stays online, marked with a retraction notice, so the scholarly record remains intact. What changes is its status: the journal is telling you that the findings, in whole or in part, no longer stand.

Retraction sits at the severe end of a spectrum of editorial actions, and it helps to know the other two:

• Correction (an erratum or corrigendum): the paper contains an error that does not undermine its conclusions. The fix is published alongside the original, and you can usually still cite the paper.
• Expression of concern: the journal has doubts serious enough to warn readers, often while an investigation runs, but has not yet withdrawn the paper. Treat these sources with caution.
• Retraction: the journal formally withdraws the paper. Its findings should not be used as evidence for your claims.

Why papers get retracted, and why the reason matters less than you think

Papers are retracted for many reasons: data fabrication, image manipulation, plagiarism, faulty analysis, results that could not be reproduced, compromised peer review, or honest errors discovered after publication. Not every retraction implies misconduct, and retraction notices usually state the grounds.

For your thesis, though, the reason matters less than the status. Whether the authors made an honest mistake or something worse happened, a retracted finding cannot carry evidential weight in your argument. An examiner who spots one will ask what else in your bibliography you failed to verify, and that is not a conversation you want in your viva.

One nuance worth knowing: journals sometimes issue partial retractions, withdrawing a specific figure, table or analysis while the rest of the paper stands. If you meet one of these, read the notice carefully and check whether the withdrawn part is the part you relied on. When in doubt, treat the source as compromised and find supporting evidence elsewhere.

Why retracted papers keep getting cited

Retractions travel slowly, and the way most researchers manage papers makes them easy to miss.

The retraction notice lives on the publisher's site, but almost nobody revisits a paper's landing page after saving the PDF. Your local copy is frozen at the moment of download, and it will look pristine forever.

Reference managers make it worse. Most store a static snapshot of the metadata and never re-check the source, so a paper retracted in March looks identical in your library in September. And because other authors keep citing retracted work in their own publications, you can pick up a retracted source second-hand from a literature review that was written before the retraction happened.

Finally, the gap between first doubts and formal retraction can be long. Investigations take time, so a paper can be under a cloud for a considerable period while continuing to look completely normal everywhere you encounter it.

How to check a single paper for retraction

Checking one paper is quick if you know where to look. The most reliable key is the DOI, because titles get abbreviated, truncated and misspelled while DOIs stay stable.

Work through these checks in order:

• Visit the paper's page on the publisher's site. Retraction notices are usually displayed prominently above the title or abstract, and the PDF is often watermarked.
• Look up the DOI in Crossref metadata. Publishers register many retractions and corrections there as updates linked to the original record.
• Search the Retraction Watch database, which tracks retractions across publishers and fields and records the stated reason for each one.
• Run the paper through a dedicated checker that queries these sources for you, using the DOI where you have it.

Use a free retraction checker

CiteDash offers a free retraction check tool for exactly this job: paste a DOI or a title and it reports the paper's current status. It is the fastest option for one-off lookups, such as a paper a colleague sent you, something you found in a citing article, or a source you are screening before adding it to your library.

If two sources ever disagree, trust the publisher's page. It is the canonical record of the paper's status, and aggregator databases occasionally lag behind it.

Screening at the point of discovery is the cheapest place to do this. A retracted paper that never enters your library never has to be hunted down later, never contaminates a draft, and never forces you to rework an argument. Thirty seconds at import saves hours at submission.

How to check your whole thesis bibliography for retractions

One paper is easy. A thesis bibliography with a hundred or more entries is where manual checking breaks down, and it is exactly where a retracted paper is most likely to hide.

The manual route is to export your reference list, pull out every DOI, and run each one through the checks above. Budget a serious block of time and keep a record of what you verified and when, because a check is only valid on the day you run it. A source that was clean at your proposal stage may not be clean at submission.

The better route is to make retraction status part of your library rather than a separate chore. In CiteDash, papers in your project library carry retraction badges, and retracted papers are blocked from citation entirely: a withdrawn source cannot enter your bibliography at all, whether you added it last week or two years ago.

Whichever route you take, timing matters. Check when you first import a source, again before each chapter goes to your supervisor, and one final time in the week before submission when your reference list has stabilised.

What to do when a paper you cited is retracted

Finding a retraction in your bibliography is not a disaster. What matters is how you respond, and examiners tend to respect a candidate who has clearly handled it.

Start by assessing what the citation is actually doing in your text:

• Load-bearing: the retracted finding directly supports one of your claims. Remove the citation, look for alternative evidence, and be prepared to soften or rework the claim if none exists.
• Contextual: the paper is one of several sources behind a general point. Drop it and let the surviving sources carry the sentence.
• The retraction is the point: if you are discussing the retraction itself, for example in a debate about methods in your field, cite the retraction notice explicitly and make the status unmistakable to the reader.

Tell your supervisor early

If a retracted source appears in anything you have already submitted or published, tell your supervisor as soon as you find it. A short note explaining what you found, when the retraction happened, and how you have fixed your text is far better than an examiner or a reviewer finding it first.

This is also worth raising as a habit in your research group. Shared reading lists and inherited bibliographies are a common way retracted work spreads, and a group that checks together catches problems that any individual member would miss.

If the retracted source appears in a published or submitted journal article rather than a thesis draft, the stakes are higher and the fix is more formal. Your supervisor will know whether a correction to your own paper is warranted, and journals have established processes for it. The principle is the same at every level: surface it yourself, promptly, with the fix already in hand.

How CiteDash catches retractions for you

CiteDash treats retraction checking as infrastructure rather than a task you have to remember. Papers in your Library carry retraction badges, and a retracted paper cannot be cited: the block is enforced at the citation level, not left to your memory or your calendar.

This works because citations in CiteDash are database objects that resolve to real papers, never free text. Every AI claim is grounded in held full-text PDFs, and the Fact Checker verifies claims against their sources before they reach you. A source that loses its standing loses its place in your reference list along with it.

Build retraction checks into your submission routine

The best defence is rhythm. A retraction check costs minutes; discovering a retracted source during your viva costs credibility.

A workable schedule for a thesis looks like this:

• At import: check status before a paper enters your library, especially anything found through a secondary citation rather than your own search.
• Per chapter: re-check the sources in each chapter before it goes to your supervisor for review.
• Before submission: sweep the full bibliography in the final week, once your reference list has stopped changing.
• Before your viva: if months pass between submission and examination, run one more sweep so nothing surprises you in the room.

Key points

Retraction is a status change you have to look for: your PDFs and your reference manager will not warn you. Check single papers by DOI against the publisher's page, Crossref and the Retraction Watch database, or use a dedicated checker. Sweep your full bibliography at every milestone, act proportionately when you find something, and tell your supervisor early.

Retraction checking also slots naturally into a wider pre-submission review. Our guide on how to check your thesis before you submit covers the full checklist, from originality and formatting to submission readiness.