Wikimedia blog

News from the Wikimedia Foundation and about the Wikimedia movement

Multilingual Wikipedia Survey Launched

In collaboration with the the Collaborative Creativity Group at UNU-MERIT (www.merit.unu.edu), we want to invite you to take the first multilingual survey of Wikipedia readers and contributors. For the
first time, this survey will provide an overview of the Wikipedia community and how the content of Wikipedia is created, used, and perceived. We therefore encourage everyone to participate in this
survey and to fill in an online questionnaire that will be made accessible to you in the coming two weeks. We have prepared survey versions in more than 20 languages. In order to keep the traffic
manageable we have chosen a staggered approach for the surveys.

The survey is currently running in Dutch, Vietnamese, and Tamil, and we have received more than 2500 complete responses already. (We can track the responses by language, so we can choose to examine any subset we want.)

The following language versions will be launched in the coming days: Russian, Arabic, Polish, Portuguese, Greek, Esperanto, Czech, Japanese, Italian, Russian, Afrikaans, Indonesian, French, Thai,
Spanish, German, English, Chinese-simplified and Chinese-traditional.

The survey will be featured in the site-wide announcement banners of those languages.

I want to extend a BIG thank you to all the volunteers who have worked on this survey, especially all the translators. We will compile translation credits for the press release when the survey is
completed.  Thanks also to the UNU-Merit team (Rishab Aiyer Ghosh, Rüdiger Glott, Herman Pijpers, Jan Philipp Schmidt), and to Naoko Komura, who has been project managing the survey since September. And, thanks to all colleagues who have given feedback along the way.

We’ve tried to design questions that make sense. Please feel free to send any and all feedback to info(at)wikipediastudy(dot)org.

Translations have been reviewed by multiple people, but if anything is an obvious error, we will try to fix it. We will not be able to address all feedback in this first run, but we will try to learn from
it for future surveys. This one won’t be perfect, but it will tell us lots of things we’ve never been able to talk about with any degree of confidence.

Finally, a note on the coming analysis, and on privacy.

In terms of analysis, UNU-Merit will collect and analyze the data, and publish analyses of the results, available under a Creative Commons Attribution/Share-Alike License on a public website as well as in
established academic journals. Anonymized data will be published under a CC-BY license for other researchers to study.

In terms of privacy, no personally identifiable information will be released by UNU-Merit or the Wikimedia Foundation without permission of the respondents. Personally identifiable data will also only be retained for a year from closure of the survey, except for participants who provide express permission to be included in a panel for a follow-on survey.

I’m looking forward to seeing the first results, and I hope many of you will take the survey. :-)

Erik Möller
Deputy Director, Wikimedia Foundation

<

8 Responses to “Multilingual Wikipedia Survey Launched”

  1. Very nice blog.

  2. Mike Shaw says:

    Wikipedia is like Darwinian evolution or democracy or English law. It is imprecise and wasteful. It is dragged about by the forcefields of religion, politics, and cliques but it is alive. It will get there in the end. Unlike former encyclopedias which are like carvings in the cliff Wikipedia respond, updates, includes new ideas, discards new ideas, moves on.
    People have become boring on the subject that Wikipedia is inaccurate. Who cares. You should always double check your sources. Where else can you look up absolutely anything from Eva Angelina’s breast size to Tensors. When I look up something I get the overview I want. I get the quirky sort of information that we are usually looking for. If I look up Blade Runner I get the speech we all remember. Wikipedia is written as if it were the answer to these sort of questions. Its eclecticism is its joy. OK your professor might slag you off if you wrote a thesis like that but it isn’t a thesis, it is a rag bag of all the bits and pieces that each heading attracts. Long may it be so. Of course heading duplication is a problem but that can be sorted any number of ways.
    There is a worrying feeling with this survey that the management (thank god evolution had no managers) are worried about the criticism from more formal sources and are going to try and straighten things out. Please don’t. Man was an evolutionary disaster whose young are born helpless and carry practically no cognitive genetic predisposition. This has proved to be the reason that man has been able to develop beyond the simple behavioral characteristics of his species. Managers would have scrapped man long before that happened. Wikipedia the same.

  3. Jay Walsh says:

    We appreciate your concerns. The questions relating to cultural or social background of participants will ultimately help us better understand who reads Wikipedia and who edits. There is no commercial interest or motivation to ask these questions.

    The more general information we have about our volunteers, the better we can build programs and resources to support ongoing work. That might mean how we conduct outreach work to bring in new editors, or it might mean how we go about organizing events with partners or other educational organizations. This kind of research is common among non profit, volunteer oriented projects.

    We’ve had some good feedback – and thousands of completed surveys so far. Along with our partners at UNU-Merit we’ll be able to better shape surveys for future studies. We should be ready to share some of the findings in the new year.

  4. B.Brother says:

    I started the survey, willing to contribute as always. Then I was pretty shocked about the questions. How is whether or not I have a partner and my income helping Wikipedia to get better??? This clearly looks like plain datamining and I aborted the survey. It changed my mind about Wikipedia – now it looks to me like a normal profit-oriented organisation exploiting my free contribution – crowdsourcing in its 2nd worst case. I’m not going to contribute anything anymore.

  5. m says:

    link to the survey, please?

  6. Hi there, just completed the survey.

    It takes quite a long time to complete; could less be more?

    I found some questions very personal, I kept having to check the URL to make sure I’m not on some spam-site / phishing site or other; as in “is the next page going to ask for a credit card number?”.

    I don’t think you want to record in your database wikipedia username together with their salaries / gender / etc. What if one of the volunteers made a simple mistake and someone gets access to the whole database?

    Can you run it over https? Can you add some more authentication? I couldn’t see a way to contact the survey committee itself, and found this page only through a search engine.

    Thanks for all your work! — Stephan

  7. Diana Rosa Mayo says:

    Dear Erik,

    I am fluent in English, Spanish and Russian. Let me know if you need help translating the multilingual survey. It ‘d be a pleasure to volunteer.

  8. Ronnie Lowry says:

    Thanks for such a great place to begin our searches for information.

    ronnie_lwry@yahoo.com