Yeah, that was high on my list of anecdotal evidence :) Ryan Kaldari On Fri, Jul 19, 2013 at 3:30 PM, Jane Darnell <jane023(a)gmail.com> wrote:

What if we only asked the question in countries where it was culturally appropriate? Even a single data point would be more useful than none. Ryan Kaldari On Fri, Jul 19, 2013 at 4:39 PM, Dan Andreescu <dandreescu(a)wikimedia.org>wrote;wrote:

Sure, we could ask the Education program to reach out to universities or university programs related to under-represented groups. The reason I ask isn't just to stir up controversy. I've done a lot of editing related to the US Civil Rights Movement, and I've noticed that there are virtually no other Wikipedians actively working on these articles (and lots of red links). The reason I ask is because I'm wondering if this lack of content is related to demographic issue or not. It seems like a fairly analogous problem to the gender gap, and thus worth doing some research on. Ryan Kaldari On Fri, Jul 19, 2013 at 5:02 PM, Diederik van Liere &lt;dvanliere(a)wikimedia.org > Ryan: suppose you would have this information what would change? Would we > prioritize features differently? Would we stop doing certain things, start > doing things? I am having a hard time finding a use case. > > D > — > Sent from Mailbox <https://www.dropbox.com/mailbox> for iPhone > > > On Fri, Jul 19, 2013 at 7:58 PM, Ryan Kaldari &lt;rkaldari(a)wikimedia.org&gt;wrote;wrote: > >> What if we only asked the question in countries where it was culturally >> appropriate? Even a single data point would be more useful than none. >> >> Ryan Kaldari >> >> >> On Fri, Jul 19, 2013 at 4:39 PM, Dan Andreescu &lt;dandreescu(a)wikimedia.org&gt;wrote;wrote: >> >>> That's a great point Tillman, and some very interesting reading >>> material. As I was reading, I was thinking, perhaps asking "Do you >>> consider yourself to be part of a (a) minority ethnicity or (b) majority >>> ethnicity?" might suffice. It's kind of what we're after and it seems more >>> universal than the US-centric "what ethnicity are you" question. >>> >>> Dan >>> >>> >>> On Fri, Jul 19, 2013 at 6:57 PM, Tilman Bayer &lt;tbayer(a)wikimedia.org&gt;wrote;wrote: >>> >>>> With all due respect and appreciation for the highly commendable >>>> intention behind the question, it is itself a symptom of systemic >>>> bias. >>>> >>>> A while ago a community member asked me the same thing - why we don't >>>> ask for the editor's race in the WMF editor surveys. I wasn't around >>>> when the demographics part of the current editor surveys series was >>>> designed, but after looking into this topic a bit, I think the answer >>>> is simple: Because these are international surveys. >>>> >>>> Survey questions about race are vastly more common and accepted in the >>>> US (and some other countries like, afaik, the UK or Australia) than in >>>> many other countries. In much of Europe, for example, asking people >>>> about their race, or classifying them racially, is considered very >>>> offensive (cf. >>>> https://en.wikipedia.org/wiki/Race_(human_classification)#European_Union >>>> ). This is particularly true in my home country, Germany, for >>>> historical reasons - personally, I don't hold very strong views on the >>>> topic, but as a German who moved to the US last year, I can tell you >>>> that it was quite an unusual experience to be asked to state my race >>>> on an official form for the first time. >>>> >>>> What's more, even when switching from "race" to the somewhat less >>>> offensive (but even more complicated) concept of "ethnicity", it seems >>>> difficult to come up with an internationally accepted list for the >>>> purposes of a global survey, cf. >>>> https://en.wikipedia.org/wiki/Race_and_ethnicity_in_censuses . >>>> >>>> If you want to wade deeper into this morass, the following monograph >>>> has some interesting information: >>>> >>>> Jürgen H.P. Hoffmeyer-Zlotnik, Uwe Warner (2009): "Die Abfrage von >>>> 'Ethnizität' in der international vergleichenden Survey-Forschung". >>>> ISBN 3-924725-15-2 http://www.ssoar.info/ssoar/handle/document/5917 >>>> >>>> It's in German, but there is an appendix ("Anhang A1" starting on >>>> p.60) which excerpts definitions of ethnicity or migratory background >>>> from censuses of 27 countries, most of them in an English translation. >>>> >>>> They also quote (p.150) an internal working paper of the ISSP >>>> (https://en.wikipedia.org/wiki/International_Social_Survey_Programme ) >>>> which recommended to drop the "ethnic" variable from their >>>> international surveys entirely: >>>> >>>> "We do not see any chance that any revision [of the questionnaires] >>>> would give something comparable across all countries. The problem is >>>> located at the conceptual level [...]. There are even big differences >>>> for the 'developed' world. For instance, for the American perspective >>>> an internal differentiation of US citizens is crucial while a >>>> differentiation of people with passports from other countries is >>>> rather meaningless. For other countries, e.g. Germany or Ireland, the >>>> reverse is true." >>>> >>>> All that said, regarding diversity efforts that focus exclusively on >>>> the United States of America, a national editor survey in the US might >>>> not encounter the above issues. You may recall this mailing list >>>> thread from 2010: >>>> >>>> http://www.gossamer-threads.com/lists/engine?do=post_view_flat;post=215980;… >>>> (aptly titled "A question for American Wikimedians") >>>> >>>> >>>> On Fri, Jul 19, 2013 at 2:11 PM, Ryan Kaldari &lt;rkaldari(a)wikimedia.org&gt; >>>>>>> > At the recent gendergap strategy retreat the issue of racial >>>> demographics >>>> > was briefly brought up but no one had any numbers on it, so we didn't >>>> know >>>> > if it was an actual issue or not. Anecdotal evidence suggests there >>>> is also >>>> > a "racial gap" among editors, but it would be nice to have some >>>> numbers on >>>> > this to facilitate discussion. I did some digging and the only >>>> statistics I >>>> > could find were about the racial demographics of American readers >>>> > ( >>>> http://strategy.wikimedia.org/wiki/American_Wikipedia_reader_demographics >>>> ). >>>> > It seems that none of our editor surveys have asked about race, >>>> although >>>> > we've asked almost every other demographic question imaginable. >>>> > >>>> > Does anyone know of any research or statistics related to the racial >>>> > demographics of Wikipedia editors? >>>> > >>>> > If not, should we consider doing a micro-survey as was done for gender >>>> > recently? ( >>>> http://meta.wikimedia.org/wiki/Research:Gender_micro-survey) >>>> Most definitely not, unless it is geolocated to the US and maybe a >>>> select few other countries. >>>> > >>>> > Ryan Kaldari >>>> > >>>> > _______________________________________________ >>>> > Analytics mailing list >>>> > Analytics(a)lists.wikimedia.org >>>> > https://lists.wikimedia.org/mailman/listinfo/analytics >>>> >>>> >>>> >>>> -- >>>> Tilman Bayer >>>> Senior Operations Analyst (Movement Communications) >>>> Wikimedia Foundation >>>> IRC (Freenode): HaeB >>>> >>>> _______________________________________________ >>>> Analytics mailing list >>>> Analytics(a)lists.wikimedia.org >>>> https://lists.wikimedia.org/mailman/listinfo/analytics >>>> >>> >>> >>> _______________________________________________ >>> Analytics mailing list >>> Analytics(a)lists.wikimedia.org >>> https://lists.wikimedia.org/mailman/listinfo/analytics >>> >>> >> > > _______________________________________________ > Analytics mailing list > Analytics(a)lists.wikimedia.org > https://lists.wikimedia.org/mailman/listinfo/analytics > >

On Saturday, July 20, 2013, Dan Andreescu wrote: This too is potentially problematic. I lived in CNMI for a while. The local population that was the dominant group was Chamorros. There were 13 white people on the whole island I lived on and 3 African Americans out of about 3,000 people. There was a sizeable chunk of Chinese workers who worked at the casino. Yes, I was part of the minority community as a white person and, yes, I was subject to bias. (It gave me a much better appreciation for African Americans in the USA to be watched like a hawk everytime I entered a store, having people occasionally follow me around to watch me.) Some countries have radically different experiences with race. Think of Japan, or places like Korea. I have had white, female American acquaintences tell me they were mistaken for Russian prostitutes in both countries. If I was trying to get a better feel for this as a potential issue, I would do a combination of research methods and then collect them all, and try to arrive at a wider picture. This would include examining any survey research done on the existing body of literature, getting infobox data for things like race (and then tabulating that against country of origins and native language), looking at where identified "minority" editors are contributing article wise, looking at those experiences in terms of say total edit reversions on articles related to that topic. (I can remember an issue about African American women in professions categories that led to the creation of the article "African American women in politics" on English Wikipedia because there was a rationale that the failure of an article existed showed the lack of necessity for this specific category grouping.) I would also do sampling on photos from meetups. I would go through any chapter reports and see what chapters have written about this internally inside their own research or based on their own membership details. I would look at the representation of who was selected by the WMF in fundraising banners. (Indians from India is different than Indians from say the USA or Australia because of the bias inside their own country.) Then I would compile that all together. While it would not be as simple or pretty as one methodology, a mixed methods approach would more likely give a better idea of the picture when the population boundaries are clearly unknown. Sincerely, Laura Hale -- -- mobile: 635209416 twitter: purplepopple blog: ozziesport.com

On 07/20/2013 12:57 AM, Tilman Bayer wrote: In every local society (country or region) there are some questions that are asked in censuses and surveys. In Sweden, it used to be "social group", which defined group 1 as academics, group 2 as those with secondary education, and 3 as workers and farmers. Today, as almost everybody has a secondary education but Sweden has become an immigration destination, a more common division is foreign-born and foreign background (having foreign-born parents). Such questions that are included in national statistics, could also be asked to the wiki community. We could work with Eurostat to find some question that can be used throughout the European Union. My unscientific feeling is that the Swedish wiki community is the mostly the "white and nerdy" demographic, being young and having some college education themselves and Swedish-born grandparents with less education. To become more inclusive, the Swedish wiki community would need to reach out to more immigrants. But that seems to be about as hard as getting sports people to edit Wikipedia. We could certainly survey current Swedish wikipedians to find out their social status and background, just like a U.S. survey could ask about race. But how should we use the results from such a survey? Should we start outreach programs to town districts where many immigrants live, and see if the survey next year gives any different results? -- Lars Aronsson (lars(a)aronsson.se) Aronsson Datateknik - http://aronsson.se

On Sat, Jul 20, 2013 at 10:52 AM, Federico Leva (Nemo) &lt;nemowiki(a)gmail.com&gt;wrote;wrote: This actually seems a bit backwards. Why not ask existing contributors why they contribute? Especially amongst targeted populations? Develop strategies for recruitment and retention based around those answers? My gut feeling is that a lot of the responses that would be listed in a questionere are based around answers like those listed. My own experience with getting female friends to edit has been more along the lines of: 1) If I want to contribute to something, I want to either get paid or get credit, 2) I do not see why I should bother to edit. What is in it for me? This issue has actually come up much, much, much more frequently for me than the issues of visual editors. I rarely see good arguments that work towards intrinsic motivation as to why a person should contribute. I'd love to see some good videos pitching why a person should contribute to Wikipedia, Wikinews, Commons, Wiktionary, Wikibooks, Wikivoyage, Wikispecies, Wikidata. -- twitter: purplepopple blog: ozziesport.com

It's hard for me to see how asking people about their race results in a win for WMF or for the movement. First off, as others point out, race is a touchy subject. How do we ask people about their race without raising legitimate concerns among many of our respondent regarding *why* we're asking them for such highly personal information, and how we intend to use it? I feel like it's easier to provide a rationale for asking about gender (tho I know that can put people off as well). Second, race (even "ethnicity") is closely tied to national/cultural/regional contexts. I don't know how Pew does it , but if we come up with our own list of racial categories, many of them are *going to be wrong*--maybe offensively so. Alternately, if we ask respondents to state their racial identity in terms that are meaningful to them, aggregating the information becomes a logistical nightmare. Third... what *would* we do with this information? Conduct targeted outreach to encourage people in country FOO to edit, but *only* if they are members of ethnicity BAR? Fourth, what are the potential unintended consequences of collecting and releasing the data? For instance, what are the consequences if we report that 80% of editors from an ethnically-divided country are members of an ethnic minority associated with an oppressive regime? It's one thing for community members to self-identify and organize on-wiki around (impossible to verify) racial identity or race-related topics. It's another for the Wikimedia Foundation to be seen as taking a *stance* on the proper racial makeup of our volunteer community. A lot easier to take a stance on categories like gender, Global North/South etc. IMO but these are also potentially hot-button issues. We have all sorts of racial gaps. Too many to address (or even articulate) individually given the resources we have. We also have socio-economic gaps, ideological gaps, and geographical gaps that are deeply intertwined with each other and with race. Which lens do we want to use? If we do want to use race as a lever for increasing editor engagement and content quality, seems to me our best shot is to spread awareness around the major race-related gaps we know about (e.g. coverage of the US Civil Rights movement) as a way of drawing attention to the general problem that racial gaps create for our topical coverage and the POV of our content. If we want to identify potential racial gaps without asking people sensitive questions, we might start by looking at language gaps. http://omnipedia.northwestern.edu/ In other words, don't tell people what flag to rally around; suggest opportunities for them to make meaningful contributions. And please, please don't racially profile people based on photos :) - J On Sat, Jul 20, 2013 at 4:39 AM, Federico Leva (Nemo) &lt;nemowiki(a)gmail.com&gt;wrote;wrote: -- Jonathan T. Morgan Research Strategist Wikimedia Foundation

Yeah, I like the idea of making people aware of the impact of the racial gap on our content, rather than asking them about their own race. - J On Tue, Jul 23, 2013 at 7:21 AM, Dan Andreescu &lt;dandreescu(a)wikimedia.org&gt;wrote;wrote: -- Jonathan T. Morgan Research Strategist Wikimedia Foundation

On Mon, Jul 22, 2013 at 12:33 PM, Lars Aronsson &lt;lars(a)aronsson.se&gt; wrote: Yes, it can be useful because understanding motivation can be a great tool for outreach and for editor retention. You can compare the list of motivations as to why people contribute versus why people fail to contribute. There often appears a great big disconnect between the list, and I have not seen much evidence to suggest that people have explored this. Why is it that some people contribute because think free sharing of knowledge is an important activity to engage in in their free time? But others do not contribute because they think the editor is too complex? Perhaps, the motivation to overcome the editing problems is higher for some populations than others... because one set is willing to do that and it is a non-issue but for another, it is a non-starter. -- twitter: purplepopple blog: ozziesport.com