home / contact / Established on 9/1/2007 --- Please send us suggestions for sites to add to this list -- and suggestions on how we might improve the site / Tell your friends about this site: we have no ad budget and depend on happy visitors to spread the word. Please report dead links and errors / Martin R. Carbone / 5123 Don Rodolfo Drive / Carlsbad, CA 92010 / Telephone: 760-603-1910 / FAX: 760-603-1930 / Website <<http://www.alphabeticalist.com>>

Alphabeticalist Topics --- A / B / C / D / E / F / G / H / I / J-K / L / M / N / O / P / Q / R / S / T / U-V / W / X-Y-Z

Analysis of “Nature” article regarding Wikipedia and Brittanica accuracy. (NATURE | Vol 438 | 15 December 2005)

1) 50 sets of articles from Wikipedia and Brittanica were selected -- one article each were in a set.
One of the 50 sets were each sent to a relevant “expert” for a peer review. It is reported in the
Nature article that Wikipedia has 4 million articles.

2) The reviewers were not told which article of the set came from which source.

3) 42 “usable” reviews were returned. Presumably the 42 included both articles from a set.

4) The reviews were examined by Nature’s news team.

5) A total of 8 serious errors were found in the reviews: four out of 42 from Wikipedia and
four out of 42 from Brittanica. In other words, they each averged one serious error for every 10 articles.

6) Factual errors, omissions and misleading statements (non-serious errors) were also found --
but they did not rise to the level of being “serious”. There were 162 such non-serious errors by
Wikipedia and 123 by Brittanica. That is about 4 non-serious errors / article by Wikipedia and 3
non-serious errors / article by Brittanica.

7) My conclusion -- Although the sample sizes were very small, both sources are a valuable
source of starting information, assuming that the selected articles represented a random sample of all articles.

Marty Carbone