In December 2005, the journal Nature published a landmark peer review of 42 science articles from both Wikipedia and Encyclopaedia Britannica. Researchers found that Wikipedia contained 162 total factual errors, omissions, or misleading statements compared to Britannica's 123 — averaging roughly 4 per article versus 3. The study's headline conclusion, widely reported by outlets including CNET and The Guardian, was that Wikipedia was "not markedly less accurate" than Britannica, suggesting rough parity rather than Wikipedia superiority.
The one metric that favored Wikipedia was a tie: both encyclopedias had four "serious errors" each. This single data point has since been frequently cited to argue Wikipedia is as good as — or better than — Britannica, but it ignores the broader error counts where Britannica performed better. Encyclopaedia Britannica itself formally challenged the study's methodology as "fatally flawed," disputing both the article selection and how errors were classified.
Since the 2005 study, no widely cited head-to-head accuracy comparison has definitively overturned its findings. Wikipedia has grown roughly sixfold in article count since then, as noted in a PMC academic review, but size alone is not a proxy for accuracy. The study remains the most rigorous direct comparison on record, and on its own terms it does not support the claim that Wikipedia is more accurate than Britannica.