CivilServant's Wikimedia studies/Being a CivilServant liaison

From Meta, a Wikimedia project coordination wiki

CivilServant (now CAT Lab) is nonprofit that works collaboratively with online communities to test tools that make their communities thrive. In all our projects we work closely with liaisons who act as our guides and connectors and help ensure our research respects the integrity of their communities. The only qualifications for being a liaison is the ability to build consensus in your community and a passion for either research or helping your Wikipedia flourish. If that describes you, we hope you will talk to us about partnering with us as a liaison in your Wikipedia community.

The information below gives you an idea of what it means to be a liaison, what we would ask of you and what you can expect from us.

The Role of the Liaison[edit]

  • Our local guides and interpreters. Our liaisons are our guides to their community culture, advising us on all levels of outreach and communication.
  • Our partners in design. Our liaisons will help us adapt the research design to meet norms of their language Wikipedia and needs of their community.
  • Consensus builder. The liaison takes the lead in building consensus - or acceptance - of the project in the community, and advising us where not to go ahead with something.
  • If interested, the liaison will be a co-author in academic publications that result from this work.

What we ask of our liaisons[edit]

  • Work with us to plan the project. Using local knowledge, the liaison will help chart: how to build consensus, who we should work with both to help design and implement the study, how to reach out to the community; etc.
  • Be part of the research design team that adapts the study design for their language Wikipedia.
  • Be the primary communications advisor, helping us decide what and how to communicate with the community. (Note: We also work with Global Voices to help translate relevant documents, so liaisons should expect to do minimal - if any - translation.)
  • Be our connector to all necessary players and deciders.
  • Join us for research summits in with other liaisons. For most projects, we bring together liaisons from all our partnering Wikipedias to workshop and finalize our collective research designs. (Expenses paid)

Note: If the above looks like a lot of work, it doesn’t have to be just you! We have had success working with “liaison teams” before and, if it works better for you, we can discuss building a “liaison team” in your community.

What a liaison relationship would look like over time[edit]

  • Stage 1: Consensus building. Confer with CivilServant staff and members of your community on what research project makes sense for your Wikipedia and to begin to build consensus.
  • Stage 2: Designing. Join us for a research Summit to workshop the design of your Wikipedia’s study. Work with CivilServant staff and members of your community to lay the foundations for the study implementation.
  • Stage 3: Consent. With support from CivilServant, lead process to ensure project has community's consent.
  • Stage 4: Running the experiment. Mostly sit and wait as the study runs itself, but also help us troubleshoot problems that may arise.
  • Stage 5: Analysis, community debriefing and research write-ups. Help us debrief your Wikipedia in the study findings.

Our values for working with communities, which we ask you to take on as well:[edit]

  • Respecting the dignity and rights of community members
  • Openness and collaboration in our research style
  • Welcoming and learning from criticism wherever it comes from

How we work with communities:[edit]

Community awareness and consent: With every study, we hold consultations with affected communities and invite people to comment (for example, posting a discussion to the Village Pump). With every study, we follow U.S. university ethics boards requirements on informed consent, which requires that our studies have minimal risk. After each study, we report the results to affected communities and invite people to comment (as with pre-study consultations). In some cases where appropriate, we notify every individual who part of a study so they can make sense of the experience, notify us about problems, and as appropriate, opt out of their data's involvement

Data and ethics: Behavioral data (from the Wikipedia logs): We only observe data that is already part of the public record on the Wikipedia site. When we publish academic papers and datasets, we will anonymize the text of comments/edits and also any usernames, to make it harder for people to be re-identified. Survey data: We will ask people to consent to take surveys, in line with our ethics boards requirements. We consider survey data to be more sensitive than already-public information and will work with communities on a case-by-case basis to balance privacy with the transparency of our research. Privacy: When working with people who have pseudonyms, we respect your personal choices about how to be identified in which situations, or whether to be anonymous

Everyone on the CivilServant core team abides by the Global Voices code of conduct. If you agree to work with us as a liaison, we ask that you do too.