Today at the Berkman Center for Internet and Society, Eszter Hargittai and Aaron Shaw offered some fascinating new evidence on the gender gap in Wikipedia. Here’s a vizThink by Willow Brugh from the talk:

Eszter Hargittai is a professor in the Communication Studies Department and the Institute for Policy Research at Northwestern University where she heads the Web Use Project. Her research focuses on the social and policy implications of digital media with a particular interest in how differences in people’s Web-use skills influence what they do online. Aaron Shaw is an Assistant Professor in the Department of Communication Studies at Northwestern University. His research focuses on political and economic dimensions of collective action online.

Who edits and contributes to Wikipedia?, asks Aaron. Why does that matter, and why should we care? It’s one of the most popular websites online, and people use it to find information on many topics. It receives billions of monthly page views, and it’s the largest free source of information. Wikipedia is also a huge repository of volunteer labor — around 41 million hour have gone into creating it.

There is a huge gender gap in Wikipedia

Women make up an estimated 16% of Wikipedia editors worldwide, and 23% of US adult Wikipedia editors are women. Lots of research has investigated this question. So far, research has looked for answers within the community of existing contributors, with help of the WIkimedia foundation. Many of this research tries to recruit people through the website — looking at the culture of the community among those who already participate. This leads to a big limitation — with those existing studies, you can’t answer questions like, “are women showing up to edit in the first place?” and “if not, why not?”

Current research can’t answer these questions because they’re not looking outside the boundaries of the site. To address these questions, Aaron and Eszter have revisited the Wikipedia gender gap question with data from from the Web Use Project. This is research where Eszter’s has shown that women and men differ in their contributions to online content, differences that have been tied to people’s web use skills.

About the Web Skills Project

Eszter has been collecting data for years on people’s internet skills (the Web Use Project) as well as data on non-contributors. The dataset also shows people’s experiences over time, something that’s extremely unique in research data. Here’s how she collected it:

Data is drawn from an urban public university (not Northwestern) that’s very diverse in its student body. Data is collected from students who are enrolled in one required course — one with 90 different sections. The data collection was done with paper/pencil surveys. Why? If you do data collection online, you will bias the survey towards people who have better online skills. They also collected people’s postal addresses to make it possible to follow-up longer term.

In the first wave of Spring 2009, they collected data from 1,115 first-years, with an 80.5% response rate. In Wave 2 in 2010, they got data from 505 respondents, and 547 people in the summer of 2012. Eszter shows us two attentiveness questions that winnow out respondents that aren’t paying attention.

The sample is a collection of young adults (2009-2012) who were born between 1989-1990, who at least started college, and of whom 48% graduated. Many are first-generation college students, and many of them were working at the time they were students. It’s a diverse group in terms of socioeconomic status, but if course they’re all at the same university– the results might not generalize to all populations.

The Web Skills project has been collecting information on the number of Internet use years, their number of internet access locations, their weekly web hours, and an “Internet skill index.” They have also asked for people’s confidence in editing Wikipedia, and whether they were ever assigned a task in school that required them to start a new Wikipedia entry or contribute to one.

Wikipedia Use Among the Web Skills Project Participants

As expected, most of the participants had seen Wikipedia. The surveys asked further questions about contributions as well:

27.6% of all these students made any kind of contribution to Wikipedia, and if you take out the number of people required in class, the number is closer to 20%. Among them, men were more likely to have contributed to Wikipedia. Among racial and ethnic differences, a class requirement of editing Wikipedia did level the playing field of contributions. Most importantly, those who score the highest on Internet skills are very significantly more likely to contribute to Wikipedia — your Internet skill in 2009 effect your likelihood to contribute to Wikipedia three years later.

Aaron Shaw now talks about a regression model that he ran, to see if any of the survey answers are related to contributions to Wikipedia. Controlling for the variation in these measures, the three significant variables are gender, skills, and whether you were required to edit Wikipedia in a school assignment.

Aaron shows us their predictions on the interactions between gender, contributions, and skills. In the sample, women are less likely to have high Internet skills, and when they have the same skill as men, they are much less likely to edit Wikipedia.

Eszter tells us that whether it’s perceived or actual skills, the point is that it matters. Whether it’s just in your head or not, your Internet skills matter.

Takeaways

This gender gap matters, says Aaron, and that it matters especially among higher skilled users. Secondly, skills really matter; people with low skills don’t contribute to Wikipedia, statistically speaking, regardless of gender. Finally, Internet skills have long term effects– Internet skills in 2009 are predicting this behavior by 2012. Finally, things that don’t seem to matter are race/ethnicity, socioeconomic status, various Internet experience, and confidence in editing Wikipedia. We don’t know why that’s the case, says Aaron, but statistically speaking, the things that matter are skills, gender, and whether you’re required to contribute in class.

Aaron and Eszter close with the following open questions:

Why aren’t skilled women more likely to contribute?
How can Wikipedia address these general Internet skill barriers?

Questions

Conversation in the discussions helped us understand what “internet skills” meant. This is a measure of basic knowledge about how the Internet works, rather than Wikipedia skills particularly.