Reference Accuracy Study

A bibliographic error is a mistake in the citation's metadata, an incorrect author name, year, volume, issue, or page range, that can make the source hard or impossible to locate. A quotation error is more serious: it means the cited source does not actually support the claim attributed to it, whether because the claim is wrong, oversimplified, or attributed to the wrong paper. Reference accuracy studies report the two separately because bibliographic errors mainly inconvenience readers, whereas quotation errors corrupt the substantive content of the scholarly record and can propagate as misinformation.

Because verifying references is laborious, samples are often modest, and a single percentage hides the uncertainty around it. A confidence interval expresses the range within which the true error rate plausibly lies given the sample size. Mogull's recalculated quotation error rate, for instance, is about 14.5 percent with a 95 percent interval of roughly 10.5 to 18.6 percent. Reporting the interval lets readers judge whether differences between journals or fields are meaningful or within sampling noise, and prevents over-interpreting a point estimate drawn from a small sample.

A major reason is that authors frequently copy references from other papers' reference lists rather than reading and verifying the original source, so an initial misquotation can be transmitted from paper to paper. Time pressure, reliance on abstracts, and reference-manager errors compound the problem. Because the original claim is rarely re-checked, a distorted finding can become entrenched as accepted fact. This propagation dynamic is precisely why reference accuracy studies focus on quotation fidelity and why citation-checking services and careful verification at the editorial stage are valuable countermeasures.