What key groups are under-represented, and why? How has their absence affected the Wikimedia projects?

Based on some of our discussions so far about barriers and incentives, I have added some ideas in the part of Participation/Attracting new participants and retaining existing participants#Ways that Wikimedia could attract new participants or retain existing participants, and I encourage the rest of you to add or edit that section based on insights from our work - we might pull from that list for our proposals next month.

Today, however, I invite you to examine the data in Participation/Participants_of_Wikimedia_projects#Demographics and think about what groups are under-represented, and what effect that's having on wikipedia. Further material is available in the survey here.

I have also written to the authors to see if we can get direct access to the survey data so we can cross-reference across characteristics that are currently reported separately, especially to see if any of the reasons for not contributing are more prevalent among under-represented populations.

In the meantime, knowing that most of the wikipedia contributors are young men without children, and from what you know about which topics are well-covered in wikipedia and what isn't, what do you think are important effects of the demographics of our contributors? Why should we try to better include under-represented groups? What knowledge might they have that we are missing?

Just a quick comment -- Netmouse, I think it is also true that editors are primarily clustered in Europe and North America: other regions of the world are less well-represented. Which leads to some parts of the world being better-covered than others.

These links might be useful:

http://en.wikipedia.org/wiki/Wikipedia:WikiProject_Countering_systemic_bias is a wikiproject aimed at countering systemic bias, primarily I believe geographically-based.

http://zerogeography.blogspot.com/2009/11/mapping-geographies-of-wikipedia.html is a blog post by Mark Graham, a research fellow at the Oxford Internet Institute. His analysis suggests that countries like the United States are well-covered in Wikipedia, and countries in Africa are least-well-covered. Germany is very dense with coverage; Chad is least-dense. (His work is based on geo-tagged articles, but the basic premise is surely accurate.)

Here is an article Mark wrote for the Guardian, covering the same substance. http://www.guardian.co.uk/technology/2009/dec/02/wikipedia-known-unknowns-geotagging-knowledge