Hi,
Since participating in the Inspire campaign, I got interested in the question of exactly how many women would be needed on Wikipedia to close the gender gap. I ran some simulations and came up with some fairly radical numbers. For example, according to my calculations, there are so few current and new female editors that, even if every current and new active, female editor stayed active for ten years, we wouldn't close the gap.
I've posted the results https://civilsociology.wordpress.com/2015/05/31/closing-the-gender-gap-on-wikipedia-results-from-some-simulations/ to my blog. It's password protected so I can share the results and get feedback without making it pubic. You can access them by using the password "wikipedia". I'm hoping some of you with experience researching gender representation on Wikipedia would be able to catch any errors.
Thanks! Jason
Jason, thanks for your work. The problem you are trying to solve, however, is still not well-defined. Yes, we lack female editors, and yes, this probably has an adverse effect on our content. Until we understand why the women are not participating, and why when they do, they drop off more rapidly than men, it is fruitless to try to ramp up participation among women. In fact, this could worsen the situation if we manage to gain on board tons of women who leave in frustration after a few weeks or months, never to come back. We would then be damaging our chances to gain editors who could be become highly valued contributors. Other, unrelated research has shown that reversions have a tendency to drive people away very effectively, and new users have become more likely to be reverted since 2006. My suspicion is that women are affected by reversions more than men. If we think of this whole problem area as a multi-step process, then I think we need to set up something like this for every nth new user (male or female, whoever agrees to participate): 1) one-on-one interviews at start of sign-up 2) periodic checkup interviews per month 3) exit interviews at end of 3-month non-activity period.
Once we understand the issues affecting newbies better, we can implement changes (or not) that can improve our lopsided participation profile (not just for gender but for all other participation gaps as well). On the content side, there is nothing preventing us from actively and aggressively starting translation efforts to spread the female biographies we already have across more language versions of Wikipedia. Wikipedia suffers from the gendergap in academic bias and is in fact worse by definition, because Wikipedia follows academia, and does not create original research (according to policy). Notability issues (because women didn't make the grade in early dictionaries of biography) become more prominent for women, just as they do for under-privileged non-white-US groups, so the women's biographies that are already out there in some language version are probably notable enough to be translated into any other language version. Having female biographies to read in any Wikipedia category breeds the creation/addition of more biographies by encouraging a "copycat" effect. Similarly, as women tend to be more oriented towards family issues, education, and daily life, we should aggressively ramp up coverage such as round-the-world customs regarding graduation ceremonies, weddings, funerals, baby showers, etc. Also, things like clothing items and accessories, fashion trends, and cooking utensils are all notoriously under-covered on Wikipedia in all languages, whereas lots of content that is there in some language could just be translated across wikis.
It is my expectation that Wikidata will make such translation tasks trivial and building interfaces to add content through translations is a type of contribution that can attract casual new users without seeming too threatening (in terms of potentially being reverted).
Jane
On Tue, Jun 2, 2015 at 2:09 AM, Jason Radford jsradford@uchicago.edu wrote:
Hi,
Since participating in the Inspire campaign, I got interested in the question of exactly how many women would be needed on Wikipedia to close the gender gap. I ran some simulations and came up with some fairly radical numbers. For example, according to my calculations, there are so few current and new female editors that, even if every current and new active, female editor stayed active for ten years, we wouldn't close the gap.
I've posted the results https://civilsociology.wordpress.com/2015/05/31/closing-the-gender-gap-on-wikipedia-results-from-some-simulations/ to my blog. It's password protected so I can share the results and get feedback without making it pubic. You can access them by using the password "wikipedia". I'm hoping some of you with experience researching gender representation on Wikipedia would be able to catch any errors.
Thanks! Jason -- Jason Radford Doctoral Student, Sociology, University of Chicago Visiting Researcher, Lazer Lab, Northeastern University *Connect*: LinkedIn http://www.linkedin.com/in/jsradford, Twitter http://www.twitter.com/jsradford, University of Chicago http://home.uchicago.edu/%7Ejsradford/ *Play Games for Science at Volunteer Science http://www.volunteerscience.com*
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Hi Jane,
On Mon, Jun 1, 2015 at 11:51 PM, Jane Darnell jane023@gmail.com wrote:
It is my expectation that Wikidata will make such translation tasks trivial and building interfaces to add content through translations is a type of contribution that can attract casual new users without seeming too threatening (in terms of potentially being reverted).
True, with the caveat that finding missing articles across languages is not trivial. We have a first version of an algorithm ready for being tested that can recommend translation tasks across languages. You can read more about it here. [1] The Language team has developed a tool called ContentTranslation, currently as a Beta feature.[2] Whether the algortihm and the tool help the movement address the gender gap and the broader demographic disparity problems is an interesting question to look into once the two are in production and more widely accessible.
Best, Leila
[1] https://meta.wikimedia.org/wiki/Research:Increasing_article_coverage [2] https://www.mediawiki.org/wiki/Content_translation
Jane
On Tue, Jun 2, 2015 at 2:09 AM, Jason Radford jsradford@uchicago.edu wrote:
Hi,
Since participating in the Inspire campaign, I got interested in the question of exactly how many women would be needed on Wikipedia to close the gender gap. I ran some simulations and came up with some fairly radical numbers. For example, according to my calculations, there are so few current and new female editors that, even if every current and new active, female editor stayed active for ten years, we wouldn't close the gap.
I've posted the results https://civilsociology.wordpress.com/2015/05/31/closing-the-gender-gap-on-wikipedia-results-from-some-simulations/ to my blog. It's password protected so I can share the results and get feedback without making it pubic. You can access them by using the password "wikipedia". I'm hoping some of you with experience researching gender representation on Wikipedia would be able to catch any errors.
Thanks! Jason -- Jason Radford Doctoral Student, Sociology, University of Chicago Visiting Researcher, Lazer Lab, Northeastern University *Connect*: LinkedIn http://www.linkedin.com/in/jsradford, Twitter http://www.twitter.com/jsradford, University of Chicago http://home.uchicago.edu/%7Ejsradford/ *Play Games for Science at Volunteer Science http://www.volunteerscience.com*
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Hi Jane, thanks, I like your argumentation, and I do agree with your slant in general, yet
hi all, second thoughts: with so many female user contributing to 'general' topics already, why should non-female human beings not be able and willing to contribute more good content on, e.g., clothing and cooking? Maybe some gender stereotypes need to be done away with here in order to close content gaps. Answers sought :)
best, Claudia
---------- Original Message ----------- From:Jane Darnell jane023@gmail.com To:Research into Wikimedia content and communities <wiki-research- l@lists.wikimedia.org> Sent:Tue, 2 Jun 2015 08:51:57 +0200 Subject:Re: [Wiki-research-l] Gender Estimates Feedback
Jason, thanks for your work. The problem you are trying to solve, however, is still not well- defined. Yes, we lack female editors, and yes, this probably has an adverse effect on our content. Until we understand why the women are not participating, and why when they do, they drop off more rapidly than men, it is fruitless to try to ramp up participation among women. In fact, this could worsen the situation if we manage to gain on board tons of women who leave in frustration after a few weeks or months, never to come back. We would then be damaging our chances to gain editors who could be become highly valued contributors. Other, unrelated research has shown that reversions have a tendency to drive people away very effectively, and new users have become more likely to be reverted since 2006. My suspicion is that women are affected by reversions more than men. If we think of this whole problem area as a multi-step process, then I think we need to set up something like this for every nth new user (male or female, whoever agrees to participate): 1) one- on-one interviews at start of sign-up 2) periodic checkup interviews per month 3) exit interviews at end of 3-month non-activity period.
Once we understand the issues affecting newbies better, we can implement changes (or not) that can improve our lopsided participation profile (not just for gender but for all other participation gaps as well). On the content side, there is nothing preventing us from actively and aggressively starting translation efforts to spread the female biographies we already have across more language versions of Wikipedia. Wikipedia suffers from the gendergap in academic bias and is in fact worse by definition, because Wikipedia follows academia, and does not create original research (according to policy). Notability issues (because women didn't make the grade in early dictionaries of biography) become more prominent for women, just as they do for under-privileged non-white-US groups, so the women's biographies that are already out there in some language version are probably notable enough to be translated into any other language version. Having female biographies to read in any Wikipedia category breeds the creation/addition of more biographies by encouraging a "copycat" effect. Similarly, as women tend to be more oriented towards family issues, education, and daily life, we should aggressively ramp up coverage such as round-the- world customs regarding graduation ceremonies, weddings, funerals, baby showers, etc. Also, things like clothing items and accessories, fashion trends, and cooking utensils are all notoriously under-covered on Wikipedia in all languages, whereas lots of content that is there in some language could just be translated across wikis.
It is my expectation that Wikidata will make such translation tasks trivial and building interfaces to add content through translations is a type of contribution that can attract casual new users without seeming too threatening (in terms of potentially being reverted).
Jane
On Tue, Jun 2, 2015 at 2:09 AM, Jason Radford jsradford@uchicago.edu wrote:
Hi,
Since participating in the Inspire campaign, I got interested in the question of exactly how many women would be needed on Wikipedia to
close
the gender gap. I ran some simulations and came up with some fairly radical numbers. For example, according to my calculations, there are so few current and new female editors that, even if every current and new active, female editor stayed active for ten years, we wouldn't close the gap.
I've posted the results <https://civilsociology.wordpress.com/2015/05/31/closing-the-gender-
gap-on-wikipedia-results-from-some-simulations/>
to my blog. It's password protected so I can share the results and get feedback without making it pubic. You can access them by using the password "wikipedia". I'm hoping some of you with experience
researching
gender representation on Wikipedia would be able to catch any errors.
Thanks! Jason -- Jason Radford Doctoral Student, Sociology, University of Chicago Visiting Researcher, Lazer Lab, Northeastern University *Connect*: LinkedIn http://www.linkedin.com/in/jsradford, Twitter http://www.twitter.com/jsradford, University of Chicago http://home.uchicago.edu/%7Ejsradford/ *Play Games for Science at Volunteer Science http://www.volunteerscience.com*
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
------- End of Original Message -------
Claudia, thanks! To be clear, I do not think that gender-stereotyped-female-content has only been created by women. On the contrary, most of it, just like the rest of the Wikiverse family of projects, has been created by men. I do feel however, that the probability of young women contributing to (for example) articles about historical costume items such as hoop skirts (for example) is higher than that for men. Just as I feel that the probability of young men contributing to (for example) articles about digital power, such as electrical motor controllers (for example) is higher than that for women. Personally, I think the need for both articles is high and that both subjects are under-represented in the English Wikipedia. Jane
On Tue, Jun 2, 2015 at 11:04 AM, koltzenburg@w4w.net wrote:
Hi Jane, thanks, I like your argumentation, and I do agree with your slant in general, yet
hi all, second thoughts: with so many female user contributing to 'general' topics already, why should non-female human beings not be able and willing to contribute more good content on, e.g., clothing and cooking? Maybe some gender stereotypes need to be done away with here in order to close content gaps. Answers sought :)
best, Claudia
---------- Original Message ----------- From:Jane Darnell jane023@gmail.com To:Research into Wikimedia content and communities <wiki-research- l@lists.wikimedia.org> Sent:Tue, 2 Jun 2015 08:51:57 +0200 Subject:Re: [Wiki-research-l] Gender Estimates Feedback
Jason, thanks for your work. The problem you are trying to solve, however, is still not well- defined. Yes, we lack female editors, and yes, this probably has an adverse effect on our content. Until we understand why the women are not participating, and why when they do, they drop off more rapidly than men, it is fruitless to try to ramp up participation among women. In fact, this could worsen the situation if we manage to gain on board tons of women who leave in frustration after a few weeks or months, never to come back. We would then be damaging our chances to gain editors who could be become highly valued contributors. Other, unrelated research has shown that reversions have a tendency to drive people away very effectively, and new users have become more likely to be reverted since 2006. My suspicion is that women are affected by reversions more than men. If we think of this whole problem area as a multi-step process, then I think we need to set up something like this for every nth new user (male or female, whoever agrees to participate): 1) one- on-one interviews at start of sign-up 2) periodic checkup interviews per month 3) exit interviews at end of 3-month non-activity period.
Once we understand the issues affecting newbies better, we can implement changes (or not) that can improve our lopsided participation profile (not just for gender but for all other participation gaps as well). On the content side, there is nothing preventing us from actively and aggressively starting translation efforts to spread the female biographies we already have across more language versions of Wikipedia. Wikipedia suffers from the gendergap in academic bias and is in fact worse by definition, because Wikipedia follows academia, and does not create original research (according to policy). Notability issues (because women didn't make the grade in early dictionaries of biography) become more prominent for women, just as they do for under-privileged non-white-US groups, so the women's biographies that are already out there in some language version are probably notable enough to be translated into any other language version. Having female biographies to read in any Wikipedia category breeds the creation/addition of more biographies by encouraging a "copycat" effect. Similarly, as women tend to be more oriented towards family issues, education, and daily life, we should aggressively ramp up coverage such as round-the- world customs regarding graduation ceremonies, weddings, funerals, baby showers, etc. Also, things like clothing items and accessories, fashion trends, and cooking utensils are all notoriously under-covered on Wikipedia in all languages, whereas lots of content that is there in some language could just be translated across wikis.
It is my expectation that Wikidata will make such translation tasks trivial and building interfaces to add content through translations is a type of contribution that can attract casual new users without seeming too threatening (in terms of potentially being reverted).
Jane
On Tue, Jun 2, 2015 at 2:09 AM, Jason Radford jsradford@uchicago.edu wrote:
Hi,
Since participating in the Inspire campaign, I got interested in the question of exactly how many women would be needed on Wikipedia to
close
the gender gap. I ran some simulations and came up with some fairly radical numbers. For example, according to my calculations, there are
so
few current and new female editors that, even if every current and new active, female editor stayed active for ten years, we wouldn't close
the
gap.
I've posted the results <https://civilsociology.wordpress.com/2015/05/31/closing-the-gender-
gap-on-wikipedia-results-from-some-simulations/>
to my blog. It's password protected so I can share the results and get feedback without making it pubic. You can access them by using the password "wikipedia". I'm hoping some of you with experience
researching
gender representation on Wikipedia would be able to catch any errors.
Thanks! Jason -- Jason Radford Doctoral Student, Sociology, University of Chicago Visiting Researcher, Lazer Lab, Northeastern University *Connect*: LinkedIn http://www.linkedin.com/in/jsradford, Twitter http://www.twitter.com/jsradford, University of Chicago http://home.uchicago.edu/%7Ejsradford/ *Play Games for Science at Volunteer Science http://www.volunteerscience.com*
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
------- End of Original Message -------
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Low participation of women is not only a Wikipedia issue. As we know, the same pattern is found in the social and political participation areas of our "advanced" societies, but there are exceptions. I want to share some details on women in national parliaments and compare it to women presence in wikis:
Country - Year - Women (%) ------------------------------- Cuba - 2013 - 48.9% Ecuador - 2013 - 41.6% Colombia - 2014 - 19.9% Saudi Arabia - 2013 - 19.9% United States - 2014 - 19.4% Uncyclopedia - 2008 - 9.6% Wikipedia - 2011 - 8.5%
Source: http://www.ipu.org/wmn-e/classif.htm
Did you excepted this numbers? Just my two cents.
2015-06-02 2:09 GMT+02:00 Jason Radford jsradford@uchicago.edu:
Hi,
Since participating in the Inspire campaign, I got interested in the question of exactly how many women would be needed on Wikipedia to close the gender gap. I ran some simulations and came up with some fairly radical numbers. For example, according to my calculations, there are so few current and new female editors that, even if every current and new active, female editor stayed active for ten years, we wouldn't close the gap.
I've posted the results https://civilsociology.wordpress.com/2015/05/31/closing-the-gender-gap-on-wikipedia-results-from-some-simulations/ to my blog. It's password protected so I can share the results and get feedback without making it pubic. You can access them by using the password "wikipedia". I'm hoping some of you with experience researching gender representation on Wikipedia would be able to catch any errors.
Thanks! Jason -- Jason Radford Doctoral Student, Sociology, University of Chicago Visiting Researcher, Lazer Lab, Northeastern University *Connect*: LinkedIn http://www.linkedin.com/in/jsradford, Twitter http://www.twitter.com/jsradford, University of Chicago http://home.uchicago.edu/%7Ejsradford/ *Play Games for Science at Volunteer Science http://www.volunteerscience.com*
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
*expect these (sorry)
2015-06-02 20:34 GMT+02:00 Emilio J. Rodríguez-Posada emijrp@gmail.com:
Low participation of women is not only a Wikipedia issue. As we know, the same pattern is found in the social and political participation areas of our "advanced" societies, but there are exceptions. I want to share some details on women in national parliaments and compare it to women presence in wikis:
Country - Year - Women (%)
Cuba - 2013 - 48.9% Ecuador - 2013 - 41.6% Colombia - 2014 - 19.9% Saudi Arabia - 2013 - 19.9% United States - 2014 - 19.4% Uncyclopedia - 2008 - 9.6% Wikipedia - 2011 - 8.5%
Source: http://www.ipu.org/wmn-e/classif.htm
Did you excepted this numbers? Just my two cents.
2015-06-02 2:09 GMT+02:00 Jason Radford jsradford@uchicago.edu:
Hi,
Since participating in the Inspire campaign, I got interested in the question of exactly how many women would be needed on Wikipedia to close the gender gap. I ran some simulations and came up with some fairly radical numbers. For example, according to my calculations, there are so few current and new female editors that, even if every current and new active, female editor stayed active for ten years, we wouldn't close the gap.
I've posted the results https://civilsociology.wordpress.com/2015/05/31/closing-the-gender-gap-on-wikipedia-results-from-some-simulations/ to my blog. It's password protected so I can share the results and get feedback without making it pubic. You can access them by using the password "wikipedia". I'm hoping some of you with experience researching gender representation on Wikipedia would be able to catch any errors.
Thanks! Jason -- Jason Radford Doctoral Student, Sociology, University of Chicago Visiting Researcher, Lazer Lab, Northeastern University *Connect*: LinkedIn http://www.linkedin.com/in/jsradford, Twitter http://www.twitter.com/jsradford, University of Chicago http://home.uchicago.edu/%7Ejsradford/ *Play Games for Science at Volunteer Science http://www.volunteerscience.com*
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
Jason,
Thanks for this thoughtful analysis. I think you've done an excellent job laying out your reasoning.
Unfortunately, as you know well, any analysis of the gender gap and the retention of women editors necessarily involves a lot of estimates and extrapolations.
Some of that is unavoidable without conducting a new full-scale research project (such as getting an accurate, up-to-date % of women editors to replace old estimates from previous studies). But other estimates can probably be replaced with fresh data at reasonable cost, and could refine your projections.
For example, it shouldn't be too difficult to generate new editor retention numbers, like those from the editor trends study, with publicly available data. It's possible that we're retaining active editors at a different rate than we were in 2009. We're certainly retaining *new* editors at a lower rate than we were back then.
Overall, I agree that radical action is probably necessary if we want to see substantial increases in women editing. The outcome of several Inspire grants might help inform those actions: https://meta.wikimedia.org/wiki/Grants:IdeaLab/Survey_women_who_don%27t_cont...
Thanks again for sharing! Please keep us posted, Jonathan
On Mon, Jun 1, 2015 at 5:09 PM, Jason Radford jsradford@uchicago.edu wrote:
Hi,
Since participating in the Inspire campaign, I got interested in the question of exactly how many women would be needed on Wikipedia to close the gender gap. I ran some simulations and came up with some fairly radical numbers. For example, according to my calculations, there are so few current and new female editors that, even if every current and new active, female editor stayed active for ten years, we wouldn't close the gap.
I've posted the results https://civilsociology.wordpress.com/2015/05/31/closing-the-gender-gap-on-wikipedia-results-from-some-simulations/ to my blog. It's password protected so I can share the results and get feedback without making it pubic. You can access them by using the password "wikipedia". I'm hoping some of you with experience researching gender representation on Wikipedia would be able to catch any errors.
Thanks! Jason -- Jason Radford Doctoral Student, Sociology, University of Chicago Visiting Researcher, Lazer Lab, Northeastern University *Connect*: LinkedIn http://www.linkedin.com/in/jsradford, Twitter http://www.twitter.com/jsradford, University of Chicago http://home.uchicago.edu/%7Ejsradford/ *Play Games for Science at Volunteer Science http://www.volunteerscience.com*
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
wiki-research-l@lists.wikimedia.org