What was the gender(gap) of Open Knowledge Festival? Data for improving it

What is the gender gap in open knowledge community and how to improve it? This article was originally meant to be published in The Open Book, but was unfortunately left out in the final editing process. The study was facilitated by Mayo Fuster Morell from Berkman Centre for Internet and Society and developed in collaboration with the Gender & diversity stream of Open Knowledge Festival in Helsinki last year.  

This contribution to the Open Book presents an study to monitor gender at Open Knowledge Festival (OKFest). As in any other event, the monitoring of gender at OKFest aims to increase gender consciousness among OKFest community, and ultimately aims to contribute to reduce gender gap in open and technological sectors.

Gender gap refers to the “gap” in gender equality, that is, when one gender predominates more than and over the other. We acknowledge that gender equality is more complex than gender balance (that is, equal presence of male and female). However, in this study we adopt gender balance as an indicator of gender equality, also assuming that better gender balance contributes to improve gender equality.

Gender “what”? What is this about?

Gender gap is the difference between women and men involvement. To provide some data to exemplify the gender gap, according to surveys developed in Wikipedia, around 88% of Wikipedia editors are male (Source: Wikimedia Foundation Annual Editors Survey 2011); according to a EU survey, 98.5% of developers in free and open source software projects are male, while they are 70% in the software industry (Source: Gender: Integrated Report of Findings. UCAM, University of Cambridge. March 2006). This is also present in conference. According to our counting from Wikisym – International Symposium on Wikis and Open Collaboration (http://www.wikisym.org), at Wikisym there is 89% men at Symposium Committee (8 men / 1 woman), 78% men in Program Committee (54 men /12 women) and 80% of the program of speakers were men .

Most technological sectors (even if not in all) are more populated by male, and projects, communities and conference in those sectors might reproduce it. However, this is not a uneatable situation. Furthermore, in some occasions the projects, communities and events not only reproduce the gender balance in its sectors, but more importantly, results in worth gender balance percentage that do not represent the sector. Importantly, projects, communities and conference could help to improve the gender balance in its sectors. There are ways to solve the gender gap projects and conferences which are gender conscience and put in practices the methodologies than reduce gender gap are able to generate gender balanced communities. This is the case of Wikihow (a community about how to manuals) which has an almost gender balance community of contributors (Source: Women in Wikihow presentation at Wikimania 2012 (http://wikimania2012.wikimedia.org/wiki/Submissions/Women_on_wikiHow). The same could be said to academic institutions such as the Berkman center for Internet and society at Harvard University (which according to our counting from its web, the Berkman center program of fellows for 2011 and 2012 is gender balanced). This is also the case of technological conferences such Personal Democracy Forum with a speaker gender balance of 60% men 40% women according to our counting (Source: http://personaldemocracy.com/conferences/nyc/2012/archive).

Which is the methodology of this event gender balance monitoring?

This study was facilitated by Mayo Fuster Morell and developed in collaboration with the Gender & diversity stream of OKFest. The methodology was based on counting male and female presence in each of the ways to engage with a conference. We acknowledge that gender is a complex constructions that goes beyond female and male categories, such as the hermaphrodite. However, for this study we used the categorization of gender as female and male for methodological purposes. We classified people according to their names or/and apparency in pictures in the web.

This study is available at a wiki together with data on other events: http://wiki.digital-commons.net/Gender

Data on gender at OKFest 2012
Visualization provided by Michael Bauer

Mean of % of gender balance in all the categories at OKFest =  26,83%
* Core Organizing Team: 83.33% m / 16.6% w
* Advisory Board: 75% m / 25% w
* Guest Programmer Planners: 59% m / 41 % w
* Featured Speakers: 81% m / 19% w
* Speakers: 72.5% m / 27,5% w
* Participant Testimonial: 71,5% m / 28,5% w

Total people % Gender Balance
Core Organizing Team Total 6 = 5 m + 1 w 83.33% m / 16.6% w
Advisory Board Total 16 = 12 m + 4 w 75% m / 25% w
Guest Programmers Total 63 = 37 m + 26 w 59% m / 41 % w
Featured Speakers Total 26 = 21 m + 5 w 81% m / 19% w
Speakers Total 381 = 276 m + 105 w 72.5% m / 27,5% w
Gender balance per stream:
Ranking of women inclusion
1. Gender and Diversity in Openness (75% w)
2. Open Cultural Heritage (42,85% w)
3. Open Development (38,88% w)
4. Data Journalism & Data Visualization (38,09% w)
5. Transparency & Accountability (32,96 w)
6. Open Research & Education (24,24% w)
7. Business and Open Data (24% w)
8. Open Geodata (22,22% w)
9. Open Democracy & Citizen Movements (21% w)
10. Open Knowledge & Sustainability (19,35% w)
11. Open Cities (17,24% w)
12. Open Design, Hard., Man. & Making (15,6% w)
13 Open Source Software (9,09 % w)
Participant Testimonials on the web Total 14 = 10 m + 4 w 71,5% m / 28,5% w
m = men w = women

Overview and best practices insights from OKFest for improving gender at conferences

Some points for an overview of gender performance at OKFest and insights on best practices for women engagement in Internet and technology related conferences:

i)  OKFest 2012 had a gender and diversity in openness stream in its program, which was not the case in previous OKFest, and it is not common in other conferences on Internet and society issues.

ii) This OKFest had several sessions in the program on gender, more than in the program of previous OKFest years.

iii) The Gender Stream had a 75% of female and 25% men. For next occasions, it might be good to see how to better balance it and engage more male in the Gender stream. However, this seems a better % than other gender events in which around 95% are female. Even if the Gender Stream is predominantly female, this had the good effect of contributing to rate the female presence in the overall conference. Over all to have a gender stream has increased the overall gender balance of the conference. In this regard, it seems a recommend best practice to other Internet & Society related events to have gender specific activities in order to attrack women.

iv) Overall OKFest had a balance of 73.16% male and 26,83% female. In this regard, OKFest has a long process to do in order to assure gender diversity and inclusion in the conference and to keep working in that direction is something to do. However, in contrast to other Internet & society conference, the actual balance at OKFest is not the worth (conference like Wikisym are around 85% male dominated, but better cases in contrast to OKfest can also be found like Open Personally Forum with 60% male 40% gender balance). In sum, there is much to be done, but this is not the worse starting point.

v) What it seems to be more problematic is what happened in some specific stream topics at OKFest 2012. That “Open Software” stream was the worth (with 91% male 9% female) is not a surprise, neither that “Hardware, Manufactures & Making” was also bad (15,6% female). However, it is in certain degree a negative surprise that topic streams like “Open Cities” (82% male 17,24% female) and “Open Democracy & Citizen Movements” (79% male 21% female) performed so badly. Those are topics that could be said that has less challenging aspects to include women that other streams. These results open up questions such as: How can we think cities or democracy with such a poor involvement of women?. For future OKFest editions there could be done a more explicit effort for assuring that the balance in those (and all) streams to be better.

vi) Observing the relationship between guest program organization and the number of speakers per stream seems to suggest that (in some cases) involving more women in organizational tasks tend to have a higher presence of more women in the program of speakers. However, the data does not provide strong statistically significant support for this.

vii) The distribution of travel bursaries making sure it also favors women could be a way to assuring women participation.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: