This website uses essential cookies for authentication, language preferences and remembering that you dismissed this notice. No tracking or advertising cookies are used. Privacy policy

How to check whether a citation is real

By · Content reviewed

Check references against publication records, whether they appear in a submission you are reviewing or in your own draft. The goal is corroboration: find matching publication details in sources independent of the document you are checking. Work through the checklist below and keep unresolved cases separate from confirmed errors.

Search the exact title

Search the title in Crossref or your library catalogue. If an exact search fails, try the main title and an author surname. Be careful with generic titles that could identify unrelated works. Record what you searched instead of interpreting an empty result as proof of fabrication.

Resolve the DOI

Open the DOI through the DOI resolver and read the destination. It should describe the same work as your reference. A missing destination needs investigation; a destination with a different title needs an identity comparison. See DOI hijacking.

Match the authors

Compare the author names on the record with the reference. Use the metadata record and the publication itself to investigate missing names or different spelling. Preserve the published order; do not infer who wrote the reference from an author discrepancy.

Cross-check the metadata

Compare year, journal, volume, issue and pages against the publication metadata. Investigate differences against the publisher page before editing. Keep your note specific: which field differs, which source you used, and what correction you made.

Worked example: a real reference and an altered year

The Oxford Academic publisher record identifies this publication: Turing, A. M. (1950). Computing machinery and intelligence. Mind, 59(236), 433–460. DOI: 10.1093/mind/LIX.236.433.

The following is deliberately altered for this illustrative example: Turing, A. M. (2020). Computing machinery and intelligence. Mind, 59(236), 433–460. DOI: 10.1093/mind/LIX.236.433. Only the year has been changed; this is not a quotation from a submitted manuscript or a Hallucite accuracy test.

Opening the identifier still leads to the 1950 work. A working DOI therefore leaves the year discrepancy unresolved until you compare the citation with the record. The appropriate note is “year differs from publisher record; check and correct”, not a conclusion about who wrote the entry or why. Read the publication separately before deciding whether it supports a claim.

Be alert to retractions

A reference can describe a real work that has since been retracted. Crossref’s Retraction Watch documentation explains the available publication updates. Read any notice before deciding how to use the work.

When to use a tool

For a whole bibliography, use Hallucite to help identify supported publications and inspect metadata discrepancies. Review unresolved references separately: source outages and incomplete coverage do not establish fabrication. Publication matching also does not establish that the source supports your sentence. See how it works.

Check a reference list against publication records

Hallucite compares references with available publication records and shows the evidence. Review flags and unresolved entries; a result does not establish authorship or claim support. Read the methodology and limitations