I think everyone's idea is good.
You can tackle these problems from a "big data" perspective as Joseph is proposing which has the disadvantage of being possibly imprecise in its assessment of gender but can potentially scan the whole of Wikipedia providing a massive dataset. And you can look at "small data" like existing lists as others are suggesting; with this approach, the assessment of gender will be more accurate but the sample size will be quite small (inviting the question whether a larger sample would reveal something different). There is a lot of benefit of doing both. Big data (a highly quantitative approach) is good at testing for unusual patterns in the data but not so good in explaining them; doing a more qualitative approach with small data analysis is likely to yield explanations. If small data yields a theory on the more precise characterisation of a pattern, then the theory can be tested on the big data to see if appears to hold more widely. The two approaches complement and reinforce one another. Neither approach is better or worse than the other, simply different tools used in research. Studying small samples requires skills that most researchers have, which is why it tends to be more popular. Big data approaches need firstly a large dataset (and Wikipedia is a very large dataset!) and secondly a strong IT skill set to work with that data, which limits the number of researchers who can undertake that kind of investigation. If Joseph has the skills to tackle a big data investigation, I hope he will tackle it. I look forward to hearing the results from any investigation. Kerry _____ From: gendergap-boun...@lists.wikimedia.org [mailto:gendergap-boun...@lists.wikimedia.org] On Behalf Of J Hayes Sent: Tuesday, 14 April 2015 5:56 AM To: Addressing gender equity and exploring ways to increase the participationof women within Wikimedia projects. Subject: Re: [Gendergap] Article: Wikipedia trolls now vs. women architects i also collected some anecdotal data about deletions and speedy deletions of MacArthur Fellows. you could study differential deletions of them or Fellows of the Royal Society. On Mon, Apr 13, 2015 at 2:44 PM, Samuel Klein <meta...@gmail.com> wrote: Emilio - thanks for the reminder of that excellent page! On Mon, Apr 13, 2015 at 1:18 PM, Jane Darnell <jane...@gmail.com> wrote: I have a theory that it is much more difficult to create bios of females in whatever category due to the systemic academic bias aginst including women's biographies in the list of "reliable sources" mostly used in Wikipedia. I would be especially interested in comparison of male-female ration of bios in established dictionaries of biography and how these compare to Wikipedia, and of those, how many such bios were previously deleted on Wikipedia and recreated. Agreed. I think one of the most effective ways to counter this sort of systemic bias is to find dictionaries of biography & encyclopedic histories of women and digitize them / make them available to editors. Those sources often do exist, though they are less commonly known or available online. We have a decent reason for them to relicense those works, especially if we can more actively help with the digitization as a result. It would be a small but precise blow against systemic bias to say "the following areas have historical & reporting bias; so we make extra effort to find and recognize additional sources, and vary criteria in inverse proportion to that bias". Sam _______________________________________________ Gendergap mailing list Gendergap@lists.wikimedia.org To manage your subscription preferences, including unsubscribing, please visit: https://lists.wikimedia.org/mailman/listinfo/gendergap
_______________________________________________ Gendergap mailing list Gendergap@lists.wikimedia.org To manage your subscription preferences, including unsubscribing, please visit: https://lists.wikimedia.org/mailman/listinfo/gendergap
