What Is a Retracted Paper — And Why You Should Care
A retracted paper is a scholarly article that the journal or publisher has officially withdrawn after discovering serious errors, misconduct, or ethical violations. It’s not just a minor correction — a retraction means the paper’s findings are unreliable, invalid, or cannot be trusted for further research.
Here’s what most students don’t realize: citing a retracted paper is an academic integrity issue. If you include a retracted source in a research paper, thesis, or dissertation, it can undermine your entire argument. Professors and reviewers can check every citation — and they often do. A single retracted reference can raise serious questions about your research quality.
Over 62,000 retractions have been logged since the Retraction Watch Database began tracking them. That number keeps growing every year as databases expand and the community becomes more vigilant. The question isn’t whether your sources might be retracted — it’s whether you’ve checked.
- A retracted paper has been officially withdrawn by the journal and should not be cited
- The Retraction Watch Database tracks over 62,000 retractions and is the most comprehensive free resource
- Zotero can automatically flag retracted papers in your library using privacy-preserving matching
- Crossref and PubMed also display retraction markers on affected records
The Retraction Watch Database
The Retraction Watch Database is the single most comprehensive public resource for finding retractions. Maintained by the Center for Scientific Integrity, it covers retractions from all major scientific disciplines and journals worldwide.
How to Search It
As of October 2024, the database now requires search criteria — you must enter at least one search field before running a search. This change was implemented because the database has grown so large that unrestricted queries were slowing data entry to a halt.
You can search using any of these fields:
- Title — The paper’s title (use keywords rather than full titles)
- Journal — The journal where it was published
- Publisher — The publishing house
- DOI or PMID — The most reliable identifier for a specific paper
- Author affiliation — The institution where the authors work
Tip: Searching by DOI or PMID is the most precise approach. If you have the DOI of the paper you’re checking, enter it directly — the database will return exactly one result if the paper has been retracted.
When you find a retraction entry, the database shows the full reason for retraction, the date of the retraction notice, and sometimes links to the original paper and the retraction notice itself.
What the Database Includes
The database covers four categories of notices:
- Formal retractions — The publisher or journal officially withdraws the paper
- Errata — Corrections are issued, but the paper remains cited
- Expressions of concern — Editors flag potential issues but haven’t retracted yet
- Retracted with editor’s note — The paper is retracted and an editor explains why in detail. This is the strongest form of retraction — it means the journal not only withdrew the paper but also publicly documented the reasons. It’s the most informative type for researchers trying to understand why a paper was withdrawn.
Only retractions count as withdrawn. Errata and expressions of concern mean something’s wrong, but they’re not the same thing.
Tool Comparison — Which Retraction Checker Should You Use?
Different databases and tools cover retractions in different ways. Here’s how the five main detection methods compare:
| Method | Coverage | Ease of Use | Limitations | Best Use Case |
|---|---|---|---|---|
| Retraction Watch Database | Over 62,000 retractions across all journals | Requires search criteria; free web access | Only covers retractions it manually enters; missing corrections | Searching for a specific paper by title, DOI, or author |
| Zotero retraction alerts | ~75% of RWDB (items with DOI or PMID) | Automatic; checks while you work | Only flags items with DOI/PMID; doesn’t search the web directly | Ongoing research workflows and citation management |
| Crossref retraction records | All retractions indexed via DOI metadata | Requires understanding update-to fields; free API |
Only covers papers with DOIs; no full-text content | Verified DOI lookups during reference verification |
| PubMed | Biomedical retractions marked in records | Free; integrated into search results | Biomedical only; doesn’t flag all retractions consistently | Literature reviews in medicine, biology, health sciences |
| Google Scholar | Indexes retracted papers but may not flag them | Zero effort; already in your workflow | Retractions often indexed but not visibly marked; easy to miss | Quick sanity checks alongside other methods |
Bottom line: Always cross-check using at least two methods. The Retraction Watch Database catches the most retractions, but Zotero alerts save you time during active research, and Crossref/PM provide institutional verification.
Zotero’s Automatic Retraction Checking
If you use Zotero as your citation manager, you don’t need to manually check every paper for retractions. Zotero’s retraction detection feature — built in partnership with Retraction Watch — automatically checks your library against the database.
How It Works (And Why Your Privacy Is Protected)
Zotero’s approach is designed to protect your privacy while still catching retractions. Here’s the step-by-step process:
- Zotero calculates a non-unique identifier — Instead of sending your full library contents to Zotero servers, it creates a “fingerprint” that could map to hundreds of publications. This fingerprint isn’t specific enough to reveal which paper you have.
- The fingerprint is compared to retraction identifiers — Zotero servers compare this partial identifier against a list of partial identifiers from retracted publications.
- Full details are requested only for potential matches — If the partial fingerprint matches a retraction, Zotero requests the full details of all possible retractions for that fingerprint.
- Local matching flags retracted items — Zotero checks your local items against the full retraction identifiers and flags any matches.
No library contents are ever shared. The server has no way of knowing whether you have the retracted work in your library or one of hundreds or thousands of other possible matches. This same approach is used by password-checking tools (like Have I Been Pwned) to verify compromised passwords without sharing them with the server.
Zotero will flag retracted items in your library with a warning banner, and it will warn you if you try to cite a retracted paper using the Word processor plugin — even if the item was added by a co-author or no longer exists in your library.
Limitations: Currently, this feature only works for items with a DOI or PMID. This covers about 75% of the Retraction Watch Database. Zotero is working on expanding coverage for items without identifiers.
Crossref’s Retraction Records
Crossref is the primary DOI registration agency for academic publications. When a paper is retracted, the publisher sends a notification to Crossref, and Crossref adds this information to the paper’s metadata record.
What the update-to Field Means
Crossref retraction records include an update-to field that shows the original DOI being replaced or linked to the retraction notice. This field appears in the metadata whether the retraction is marked by the publisher or by Retraction Watch itself.
When you resolve a DOI at doi.org, you’re querying Crossref’s metadata directly. If the paper has been retracted, the resolved page will usually include a retraction notice link or a “Retracted” status indicator. This makes Crossref a fast verification method — a 10-second check at doi.org can confirm whether a specific paper has been officially withdrawn.
Key detail: Crossref metadata can be queried directly using the REST API at api.crossref.org/works/{doi} without any sign-up. You can paste a DOI directly into your browser’s address bar and receive JSON metadata showing the retraction status.
PubMed and Other Database Retraction Markers
PubMed
PUBMED maintains 35+ million biomedical citations. When a biomedical paper is retracted, PubMed marks it with a retraction notice in the record. You can see the retraction flag in the article’s “PubMeC” or “Comments & Corrections” section.
Limitations: PubMed only covers the biomedical and life sciences. It doesn’t flag retractions for papers in engineering, social sciences, or other disciplines.
Google Scholar
Google Scholar indexes retractions from most major publishers, but it doesn’t consistently flag them. A retracted paper may still appear in search results without any visible warning. This makes Google Scholar useful for a quick sanity check, but not reliable as a standalone retraction checker.
The common mistake students make is assuming a paper is safe just because they found it on Google Scholar. A paper can be retracted and still indexed — the retraction may not appear in the search snippet, and you’d need to click through to the journal’s site to verify status.
What to Do When You Cite a Retracted Paper
If you discover that one of your sources has been retracted, don’t panic. Here’s what to do:
- Replace the source — Find a non-retracted paper that covers the same material. Use Google Scholar, PubMed, or your university library database to search for alternative sources.
- If no alternative exists — If the retracted paper was the only source on a specific topic, note in your methodology that you’re using withdrawn data and explain why. Be transparent with your professor.
- If you’re using Zotero — The retraction warning will flag the item. Remove it from your bibliography and replace it with a verified alternative.
- If you already cited it — If you’ve already submitted a paper with a retracted citation, inform your professor immediately. Most professors will appreciate the honesty and will give you a chance to replace the source.
Remember: Transparency is always better than hiding a mistake. Professors expect researchers to verify sources — and if you discover a problem after the fact, addressing it honestly is the professional thing to do.
Retraction Checking Workflow — Visual Guide
Here’s a quick-reference workflow you can use as a checklist when verifying any source. Think of it as a decision tree:
Step 1: Start with the paper’s DOI
- If the paper has a DOI, resolve it at
doi.organd check the page for retraction notices - If no DOI, note the title and journal for the next step
Step 2: Search the Retraction Watch Database
- Enter the DOI, title keywords, or journal name
- Check if the paper appears in the database (remember: search criteria are now required)
Step 3: Check your Zotero library
- If you use Zotero, the retraction alerts will flag any retracted items automatically
- Look for the yellow warning banner next to retracted items
Step 4: Verify with Crossref or PubMed
- Crossref metadata will show
update-tofields for retractions - PubMed marks biomedical retractions in article records
Step 5: Decide
- If retracted — Replace the source with a non-retracted alternative
- If not retracted — Safe to cite
- If uncertain — Cross-check using the next method and document your verification
Retraction checking workflow: a five-step decision tree for verifying sources
This visual guide replaces the need to read through all five methods. When you’re verifying sources, follow the steps in order — they’re designed so that if Step 2 finds a retraction, you don’t need to continue to Steps 3 and 4. The workflow prioritizes speed without sacrificing thoroughness.
Frequently Asked Questions
Can a paper be “de-retracted”?
Yes, but it’s extremely rare. A paper can only be reinstated if the retraction was based on a procedural error (like a wrong DOI assignment) rather than scientific misconduct. This almost never happens.
What’s the difference between a retraction and a correction?
A correction (or erratum) means there’s an error in the paper — a typo, calculation mistake, or missing author — but the paper itself is still valid. A retraction means the paper’s conclusions are fundamentally unreliable.
Do retractions affect citations in other papers?
Yes. Once a paper is retracted, it should not be cited by any subsequent research. Every paper citing it should be reviewed to replace the retracted source.
Is checking for retractions required by universities?
Most universities don’t explicitly require retraction checking, but academic integrity policies expect you to verify your sources. Citing a retracted paper can be treated as evidence of poor research practices.
Conclusion
Checking if a paper is retracted is one of the most important verification steps in academic research. You don’t need specialized tools or subscriptions — the Retraction Watch Database, Zotero alerts, and Crossref metadata all provide free access to retraction information.
Always cross-check at least two methods before citing a paper. The effort takes less than a minute per source and protects you from the academic consequences of citing unreliable research.
For more guidance on verifying sources, check the annotated bibliography guide for citation best practices and our best academic writing tools guide for citation management recommendations.