Wikimedia Fellowships/Project Ideas/Editor engagement strategies for new or smaller Wikipedias

This fellowship would take a data-driven approach to research, experiment, and develop best practices for editor engagement and retention, focusing on the specific needs of “smaller” (less than 50 highly active editors per month, say) editing communities.

Background edit

Projects that have comparatively small numbers of active editors (this includes both newer Wikipedias and those with a relatively small number of speakers of the language world-wide) have unique problems and require unique solutions to encourage new editors and retain them. Although many new Wikipedias look to the more established language versions (English, German, etc) when developing policies and procedures, many of which are directly translated into the new language, these larger, late-stage projects may not be the best model for a community with only a few hundred (or less) editors.

For example, vandalism is a large concern for English Wikipedia, and in order to avoid welcoming lots of vandals, the community does not send automated welcomes to all new editors. However, vandalism is not a major concern for the Bengali Wikipedia at this point and they would like to welcome all new editors and really emphasize “be bold”, so automated welcoming makes more sense for this project. The Bengali Wikipedia is finding that automated welcoming of all editors may be a good way to increase contributions, and there are probably other options for increased “friendliness” at several points along the new user path that could be tailored to the needs of this size community. The reality about smaller projects is there are not many information on the editing environment and their activities. As well as they have problem with initiating outreach for their projects.

A fellowship project focused on gathering information, running tests, measuring and sharing outcomes, and developing best practices of on-wiki initiatives aimed at new editors for a given set of smaller Wikipedias could help improve editor retention and grow these communities.

Goals edit

Listing the “smaller” Wikipedias.[1] Gathering data like, number of small Wikipedia and the number of total speakers world-wide. As well as the number of highly active editors and other statistical data. Collecting information on how the small Wikipedias are trying to attract more contributors, what kind of initiatives they have taken to improve editor retention.

Identifying the obstacles to editor retention in small Wikipedias. For example, finding out the reasons why new editors don’t feel comfortable to contribute, what kind of editing environment and interface they expect, etc. Running test projects on different communities of different regions to find out the social and cultural impact/expectation in contributing to Wikipedia (if they have any) etc.

Improving overall editor retention among new editors in small Wikipedias.

Creating the basic editing infrastructure (i.e., basic guidelines pages) in a group of “small” Wikipedias to help new editors. Which may include:

Establishing a basic standards of an interactive and effective welcome message.
Creating the list of basic help and guidelines pages (e.g., guidelines on notability, copyright etc) to prepare. These help and guidelines pages will be easy and less time consuming to read, only emphasize on the core principles of Wikipedia, common through all Wikipedias. Those will be written in plain English and will be translated to other languages later on, and will be imported to local Wikipedias.
Establishing a trained and skilled regular welcome committee/group of mentors/helpers who will keep in touch with new editors and will guide them.

Creating a basic page to ask questions where they can ask freely and easily. Need to make sure it will be helpful even to those who do not know how to edit Wikipedia. The page should be appealing and interactive to new editors.

Projects edit

Phase 1 – Information collection, feedback (2–3 months) edit

Identifying the number of “smaller” Wikipedia,[1] performing quantitative and qualitative research to define barriers to growth of active editors, as well as assembling the other necessary information. This will be done with the help of statistical data from the software and thorough editor interviews. During this time following information might be collected.

  • Number of small Wikipedias with the number of total speakers worldwide;
  • Number of registered users, active edtiors and highly active edtiors in each small Wikipedias;
  • Number of established users (users with special rights like, administrator, rollbacker, reviewer, eliminator, autopatrolled etc) in smaller Wikipedias;
  • Rate of new article creation;
  • Number of total edits in each day;
  • Number of edits on talk pages and user talk pages (excluding bot edits).

Interviewing several established users in potential wiki communities to collect the information on:

  • Basic needs of the community;
  • Technical difficulties & restrictions of using the Unicode for the language. (OS support, Browser Support, Keyboards, etc.)
  • Expectation of new users/editors;
  • Affect of bureaucracy on new editors;
  • Obstacles to editor retention in a community;
  • Status of the new editors’ automatic welcome message (not on talk page), and welcome message (on talk page);
  • Finding out if they have any idea like automatic welcome/mentorship/helper groups etc;
  • Status of using IP-welcome;
  • Individual initiatives taken by the community or preferences on editor retention;
  • Status of basic help and policy pages;
  • Finding out if they have any prepared model for new editors to work on;
  • Use of barnstars and other methods to encourage new editors;
  • Concerns on vandalism.
  • Idea of payed edits by NGO or any other entity.

Phase 2 – Research, designing, and build-up the pilot projects (2–4 months) edit

Processing the statistical data and feedback from interviews (e.g., individual editor retention initiatives in other Wikipedias) and designing a pilot project to run on specific Wikipedias of different regions.[2] In the mean time preparing basic infrastructure for all smaller Wikipedias.

Following steps should be done for all small Wikipedias[3]:

  • Designing interactive welcome messages which will be short, fulfill the needs of new users, will make a good impression of the project (like the linked pages should be well-established) and getting them localized and modified as per community needs;
  • Designing necessary help, guideline, and tutorial pages and preparing the translation;
  • Creating a central question/help pages where new users can ask question easily, as well as a FAQ page to answer most common questions;

Following steps should be done for those Wikipedias on pilot projects:

  • Establishing a welcome committee or group of mentors consisting experienced Wikipedians;
  • Trained them to make them capable to provide effective help to new editors (e.g., how to answer to new editors, how to make request to them, how to keep in touch with them, etc);
  • Outreach program is important to get the maximum coverage (using the visitors of the sites with central notice) on editor retention during the pilot projects. This can be both informational and motivational. The concept of outreach here is not only about informing the potential helpers to get involved in the project, but also encouraging visitors to be bold to edit Wikipedia. Many small Wikipedias are the largest online encyclopedia in their languages. These facts can be pointed visitors to attract new editors to work on their Wikipedias, and to built an enriched encyclopedia in their languages. Creating a page with attractive motivational texts to motivate new editors to contribute Wikipedia will be useful.

Bureaucracy is one of the major concerns against editor retention in smaller projects. In order to get rid of unnecessary bureaucracy on small projects, we can encourage established users to encourage newbies with barnstars and appreciation messages etc. WikiLove extension can be useful send notices and other forms of appreciation to appreciate the work of potential new editors. It is also important to contact to new users in friendly and less formal way and encourage them to be bold.

Phase 3 – Running the pilot projects (3–4 months) edit

Examine new editor engagement strategies of a group of “smaller” Wikipedias – e.g., automatic welcoming, mentorship, targeted welcoming committees, etc — and compare their effectiveness across projects with similar editor populations and article count. The pilot projects will include projects from different language communities from different regions and will run in different ways which may include:

  • Conduct testing of community interface changes (welcome messages (automated vs non, number of links, etc), user warnings, and so on) across multiple projects and measure change to editor retention;
  • Set up welcoming/mentoring committees in projects lacking them and test efficacy of human welcoming versus bot or no contact;
  • Localize the Feedback Dashboard in a couple of “smaller” Wikipedias and gauge impact on new editor retention;
  • Running the pilot projects with site notices in some projects and without in the rest to measure the effectiveness of using site notices on this.

The timespan of this phase may vary.[4]

Phase 4 – Overall review, measurements and reporting (2 months) edit

Findings from these experiments could be synthesized into best practices and shared with relevant communities, incorporated into the incubator process, etc. The metrics of measurements[5] are may include the following:

Quantitative metrics
  • Welcome messages sent per day;
  • Visitors per day;
  • Visits per week per visitor;
  • Active conversations per day (measured via page views and unique edits);
  • Count/average of new editors interacted with per mentor per week.
Qualitative metrics
  • Examples of direct mentorship/coaching;
  • Feedback participation on feedback dashboard;
  • Interviewing mentors and established users on how the project is going.
Comparative metrics
  • Edit count by namespace;
  • Bytes added;
  • Logins and editing sessions per day/week/month;
  • User rights gained;
  • Blocks of new editors;
  • Number of reverted edits (to find out the vandalism rate);
  • Number of articles created;
  • Number of deletions of created articles;
  • Collaborations among users (edits on talk pages, articles, etc);
  • Warning templates posted per user page.

Phase 5 – Implementation (1 month) edit

After rectifying the errors and developing the ideas based on experiences of pilot projects in different communities, an open call for adopting the successful ideas in the smaller Wikipedias will be offered. Implementation phase may include:

  • Preparing complete documentation on the process;
  • Creating a group of fellow Wikipedians who will help interested communities to implemented the idea also to make necessary changes for them.

Fellow(s) edit

Qualifications for a fellow leading a project like this might include:

  • Ability to work across linguistic/project lines, engaging with editors from different communities
  • Knowledge of basic Wikipedia infrastructure — where to find key policies/projects, how to reach out to specific kinds of editors (vandal fighters, welcomers, etc.)
  • Basic knowledge of qualitative/quantitative research methods (e.g., with help from WMF staff, setting up A/B tests, comparing samples, conducting interviews, qualitative coding of edits and users)
  • Ability to share findings from research and tests and develop them into best practices

Rationale edit

Targeted – improving editor retention is a strategic priority for 2012.
Actionable – the focus would be on experiments and pilots rather than general research.
Impactful – the project should focus not just on one language Wikipedia, but several, and demonstrate increased numbers of active editors.
Sustainable – experiments and data-driven research would be used to develop some best practices and share this knowledge among all volunteers working in the space Scalable – see Impactful and Sustainable.
Measurable – the approach would be data-driven by design, experiment outcomes would show measurable increases in numbers of active editors and edit counts due to changes in welcoming, etc

Submitted by edit

Siko and Tanvir. Anyone is welcome to help contribute, edit, and shape this idea into something that would be valuable to the communities it is intended to serve.

Notes edit

  1. a b Identifying all the Wikipedias and gathering preliminary information on current status is important for future interests, as well as to make sure we do the project thoroughly and leave no projects behind.
  2. Indic languages Wikipedias, Southeast Asian languages Wikipedias, and African languages Wikipedias will probably be the best choice to run the pilot projects as they hold significant number of users in their sides and global importance. As well as they are culturally diverse, that is useful to judge the cultural impact on Wikipedias.
  3. It is not like it will be exported to all small Wikipedias right away. It is that, the templates and pages will be created as they are “usable” to most of the small Wikipedias, so after finishing the project or anytime a small wiki wish to use it, they can just copy that over to their wikis. There is an implementation phase to help on preparing necessary documentation on implementation.
  4. The time to run the pilot projects can increase because new editors often take some time to start their first edits, so it is very possible that the process will be slow at the start-up. That is where outreach programs help to let people know. As pilot will run in several projects at a time, duration can vary project-wise.
  5. These metrics are mostly from Teahose projects, as both work on editor engagement.

Endorsements edit

This section is for endorsements by Wikimedia community volunteers. Please note that this is not a debate, vote, or poll, but is rather a space for volunteers to describe in detail why they think a project idea is of value. If you have concerns or questions rather than an endorsement to make, please use the idea Talk page. Endorsements by volunteers willing to work in collaboration with a fellowship recipient on a project are highly encouraged.

  • I love this idea, and would like to see the outcomes of this fellowship, hoping that some of the findings will be from Africa-language Wikipedias. Abbasjnr (talk) 12:10, 19 February 2012 (UTC)[reply]
  • I agree with Abbas here. put it in place for high potential language versions in south east asia, preferably not with focus on India as a state to avoid partly double work with the wmf india program, and africa, swahili would be great, to gain maximal impact and measurable results. more details on methodological aspects down the road would be fine but taking the cultural aspects into account, as proposed, looks well thought through and therefore I endorse the proposal as such, regards --Jan eissfeldt (talk) 15:30, 22 February 2012 (UTC)[reply]
  • I support this idea. --mzito (talk) 03:31, 25 February 2012 (UTC)[reply]
  • Great idea to bring together people with knowledge about small languages in areas where the people have a totally different way of life. They need more support and some of us can give it to them. ZeaForUs (talk)
  • Really a great idea. Singhalawap (talk) 17:18, 28 March 2012 (UTC)[reply]
  • As a Wikimedia steward I often work with smaller projects and have to deal with their problems. The proposed project affords a good opportunity to evaluate these problems on a professional base and solve them. Thus, I strongly endorse this idea. Cheers, —DerHexer (Talk) 19:01, 17 April 2012 (UTC)[reply]