A Paper You Cited Was Retracted: What to Do Next

A retraction after your citation is a routine research event

Papers get retracted for many reasons: honest error, data problems, plagiarism, image manipulation, or fabrication. Retraction Watch, now integrated into Crossref’s API, tracks more than 65,000 retractions across the academic record (Retraction Watch database, 2026). If you cite ten sources in a chapter, the probability that at least one will be retracted at some point in the next decade is real.

What to do if a paper you cited is retracted is not a crisis question. It is a triage question. Citing a paper that was later retracted is not misconduct on your part; failing to update the citation once you know is a different matter. This piece walks the four decisions you have to make, in order.

Why this matters

Examiners spot-check bibliographies during their read of the thesis (Park, 2003). A retracted paper that appears in your reference list without a retraction note reads as either carelessness or ignorance of the retraction. Neither is a fatal impression, but both are avoidable. The candidate who caught the retraction, evaluated its implications, and updated the citation is a candidate the panel trusts on the rest of the bibliography.

There is also a real intellectual issue. If a paper is retracted because the underlying finding did not hold, and your argument relied on that finding, the argument now needs re-examination. This is not the retraction’s fault; it is what happens when the evidence base changes under a citation. The Thesisroom method treats verification as a whole-degree habit, so retractions surface while there is time to respond.

Decision 1: Confirm the retraction and read the notice

The first decision is a factual one. Search Retraction Watch (https://retractionwatch.com) for the paper title or DOI, or query the Crossref API for the retraction flag. If the paper is retracted, the notice will name the retracting parties (authors, editors, or both) and give the reason.

Retraction reasons matter. A retraction for image manipulation in a figure you did not cite is different from a retraction because the central finding did not replicate. The Committee on Publication Ethics (COPE) publishes guidance for citing retracted papers: the citation should be clearly marked as retracted, and the retraction notice itself should be cited with its own DOI (COPE, 2019).

Decision 2: Decide whether the retraction affects your argument

The second decision is intellectual. Locate every place in your thesis where the retracted paper is cited. For each, ask a specific question: does my argument rely on the finding this paper reported? Three answers are possible.

If the paper is cited for background or historical context, the retraction usually does not damage your argument. You keep the citation and add the retraction note.

If the paper is cited for a specific finding that supported one of your claims, and the retraction reason bears on that finding, the argument needs adjustment. Either find a replacement citation from a paper that reports the same finding (or a closely related one), or reframe your claim to fit the evidence that remains.

If the paper is cited for a methodological choice you made, the retraction may or may not affect your choice. A methodology that was validated in the retracted paper needs a fresh source of validation. A methodology that was widely used before that paper is not undermined by its retraction.

Decision 3: Update the citation the way COPE recommends

The third decision is bibliographic. COPE’s retraction guidelines (COPE, 2019) recommend a specific handling. When a retracted paper is cited, the reference list entry is prefixed with “RETRACTED:” or an equivalent marker, and the citation includes both the original DOI and the retraction notice DOI. The reader should be able to see, from the reference list alone, that the paper was retracted.

In practice, this looks like:

Author, A. (2019). Title of paper. Journal Name, 12(3), 456-478. RETRACTED. Retraction notice: https://doi.org/[retraction-DOI]

In the body of the thesis, the citation should note the retraction where it appears. “Author (2019, since retracted) reported that...” is honest, defensible, and preserves the reasoning while making the retraction visible.

Decision 4: Decide whether to keep the citation at all

The fourth decision is whether the citation belongs in the final thesis. There are cases where the answer is no. A retracted paper cited only for a historical or specific-empirical claim, where the retraction reason directly undermines the claim, is often better dropped. If the same claim can be supported by a different, unretracted paper, replace the citation. Thesisroom’s source finder is built for exactly this replacement task: enter the claim, and it returns verified papers that support it.

Do not silently remove the citation if the retraction happened after you built your argument on it. The intellectual honest move, if the argument still holds under a different citation, is to change the citation and the wording so the reader can see the source of the claim. If the argument no longer holds, reframe the claim to fit what the evidence still supports.

The specific problem no reference manager fixes

Reference managers store references. They do not re-verify them. A paper you saved to Zotero or Mendeley in 2023 and cited in 2024 will not be flagged in your library if it is retracted in 2026. The retraction happens in the world; your reference manager does not know.

This is the specific gap the Thesisroom living bibliography fills. It re-checks the bibliography on an ongoing basis against Retraction Watch and Crossref. When a paper you have cited is retracted, the entry is flagged as newly retracted, with the retraction notice and the reason. A doctoral candidate who runs a monthly check catches retractions while there is still time to respond. A candidate who catches the retraction the week before submission has fewer options and less time to think.

What Thesisroom does with this

Thesisroom’s citation guard looks the bibliography up in the academic record at any stage of the degree. Retracted papers are surfaced with their retraction notice and reason. The living bibliography, part of the Monthly lifecycle plan, keeps the check running so retractions that happen after you cite are caught while you can still act.

The tool does not decide whether a retraction damages your argument. That is a reading task the writer has to do. It does surface the retraction and the reason, so the reading is targeted at the citations that need attention.

Frequently asked questions

Am I in trouble for citing a paper that was later retracted?

No. Citing a paper in good faith before it was retracted is not misconduct, and examiners understand this. What matters is what you do once you know. Updating the citation, marking the retraction, and re-checking whether your argument still holds is the expected response.

Do I need to remove every retracted citation from my thesis?

Not necessarily. Retracted papers can still be cited for historical or contextual reasons, as long as the citation is clearly marked as retracted. What has to happen is the reader has to be able to see, from the citation itself, that the paper was retracted. COPE (2019) publishes the recommended format for this.

How do I check if a paper I cited has been retracted?

Search Retraction Watch (https://retractionwatch.com) for the title or DOI, or query the Crossref API. Since September 2023, Crossref includes retraction status in its metadata responses. A one-off manual check is slow at bibliography scale; Citation Check runs the check across the whole reference list in one pass.

What if my argument depended on the retracted finding?

Reframe the argument to fit the evidence that still stands. If the finding was central and no other source supports it, the argument needs rework, not a citation swap. This is one reason to verify sources during the literature review, so the argument is built on a foundation that has already been checked.

Do reference managers catch retractions?

No. Zotero, Mendeley, and EndNote store references but do not re-verify them against retraction databases. A paper you saved before it was retracted stays as-is in your library. This is the specific gap the living bibliography fills: it re-checks entries over time and flags new retractions.

How often should I re-check my bibliography for retractions?

At least once a year during a multi-year doctorate, and again before submission. Papers are retracted throughout the year, and the retraction rate has risen sharply in recent years (Retraction Watch database, 2026). Running the check monthly is easier once the living bibliography is set up, because the tool does the re-check for you.

One takeaway before you close this tab

Retractions after citation are routine, not catastrophic. The four decisions above cover what to do: confirm the retraction, decide whether it damages the argument, update the citation the way COPE recommends, and decide whether to keep it at all. When you want ongoing re-checking of the whole bibliography, Thesisroom’s citation guard surfaces new retractions as they land, so you have time to respond.

References

Committee on Publication Ethics (COPE). (2019). Retraction Guidelines. Version 3.

Park, C. (2003). Levelling the Playing Field: Towards Best Practice in the Doctoral Viva. Higher Education Review, 36(1), 47-67.

Retraction Watch. (2026). Retraction Watch Database. https://retractionwatch.com