The Wikipedia Library/1Lib1Ref/Lessons/May 2019

The Wikipedia Library

#1Lib1Ref Year 4
May 2019 Report

The May campaign this year saw an increase compared to the previous year in the total number of edits made with the hashtag by 70% and about a 100% increase in the number of events organised globally. The campaign saw major additions, new entrants (languages and communities), and a new sense of drive to attain the goals of the initiative. The May campaign reached appreciable levels of participation despite only running in its second year. It also saw new communities that had never participated in #1Lib1Ref before and extensive participation from emerging communities and languages. The diversity in language of participation exceeds the January campaign given the size and scale of both campaigns. The campaign also saw new language communities participating in their first ever #1Lib1Ref. Even though the English Wikipedia took the lead with about 44% of the total number of contributions made during the campaign, the Farsi Wikipedia (which participated for the first time in any #1Lib1Ref activity) contributed about 22% of the total.

As a campaign #1Lib1Ref provides a scalable method of supporting outreach in diverse communities and allows adaptation to ensure localisation and ownership. The impact of the campaign has been generally felt this year through its activation of old inactive contributors, on-boarding of new contributors and organizers, and the participation of affiliates that lead the campaign for the first time.

Organization

In the run-up to the January iteration a lot was done around researching ways to improve the campaign. This lead to the revamping of the campaign landing page, creation of resources, and standardization of documentation to ensure the campaign was easily adaptable and scalable.

We improved our strategy for supporting the coordinating committee this year. During the January campaign the Global Team were given a clear set of goals and roles but were left to contribute in their own capacities without regularly meeting to check in on efforts and (or) clearly instructing them during the campaign period. Conversely, during the May campaign, a small section of the global team was regularly engaged and checked on; they were also clearly assigned to specific regions. This led to easy reporting and clear mapping of who was doing or had achieved what. This seemed to work, even though on a small scale; it will be key to test the same approach during the January campaign in 2020.

The TWL team also encouraged the adoption of webinars and supported some of these by being present and providing insights. At least four were organized during the May campaign and these were extremely effective, as in countries like South Africa the webinars had secondary effects of four previously unplanned offline or in-person activities being organised. We are going to investigate supporting more webinars prior to and during the January campaign.

As the campaign enters its fifth year we wish to have a robust campaign system that runs on very little fuel from the TWL team; as described above the global team has already gained some momentum and hence provide us ample time to build partnerships to extend our reach into regions where the program/campaign or even Wikipedia has had no impact or reach. We established a partnership with AfroCrowd as they were at the time exploring options of establishing a community in the target region. The alliance with AfroCrowd during the May campaign was a pilot of how TWL can work with entities within the movement to achieve collective impact.

Learning resources

We focused this year on making the campaign more sustainable and on making resources that help the team support new participants/communities/regions. In order to further attain these objectives, an additional resource was created to facilitate the use of the dashboard. In the feedback on the January campaign, we identified that we needed to create Dashboard Resources. To further ensure that participants are well prepared for the campaign it was important to provide definite dates for when the #1Lib1Ref campaigns run (in January & May) and to provide further insights about the ability of organizers to obtain support year-round.

Tools

The Hashtags tool has and continues to be our primary means of tracking campaign-related edits across the projects. Even though our metrics are heavily based on the tool we try to collate results from both the hashtag tool and the Programs & Events Dashboard. The growing use of the hashtags tool not only warranted an improvement earlier this year but a continuous effort to add on features that make its data more meaningful.

The Dashboard is also growing to become a very significant component of the campaign as it provides data on certain metrics (words added, number of events organised, etc.) that the hashtag tool is unable to look at. This has generated several requests or suggestions for improvement. The TWL team has been meeting with the Wiki Education team that maintains the tool for adding further functions in the coming months.

Community Engagement

Following the findings from the January campaign, the TWL team took a deliberate attempt to experiment with new strategies to help develop the following:

Improving engagement with the Global Team
Including communities who have never participated in the campaign. There was at least one new participating group from each of Asia, Africa and Latin America.
Targeting communities that have been left out of our movement - There was also a deliberate focus to use the campaign to reach out new regions to our movement, and the Caribbean was the target for this May.

Eighteen press mentions of the campaign were documented. Over 1.55 million people were reached on social media by 1,229 posts made by 566 users, and campaign-related Tweets from 39 countries had over 3.7 million impressions.

Stats

The final hashtags tally was 2,384 edits in 24 languages, including first-place English Wikipedia with 1037 edits. In more detail:

2.82 million words were added as recorded on the Dashboard
527 editors participated
1,190 articles were created and 9,511 total were edited
13 wikis saw more than 5 edits, 6 more than 50, and 2 more than 500

Evaluation

Even though the English Wikipedia topped the campaign in terms of the number of contributions (44%), the Farsi Wikipedia was not far behind with contributions at 22% of the total number of contributions, a very good achievement for a first time. The Farsi Wikipedia was the biggest beneficiary of the total number of reactivated users (61 reactivated accounts) while the Central Bikol Wikipedia took advantage of the campaign to massively recruit new editors (49 participants). The number of new editors from the Central Bikol Wikipedia exceeded 90% of the total number of contributors from that language Wikipedia. The situation was not very different from other language Wikipedias, as new users constituted more than 60% of the total number of contributors on average. This stat affirms the earlier proposition that 1lib1ref could be the ultimate onboarding campaign within our movement.

Future Plans

The campaign has been adopted as a year-round activity with two major campaigns in January and May this year. We encourage this for future campaigns as it provides a means for those who cannot still make the mainstream campaigns to have an opportunity. This new development and support has been highlighted on the site to ensure awareness however we are aware that it needs to be spoken about more openly.
We want to explore a better strategy for working with the WMF grants team to meet the needs of communities working around #1Lib1Ref campaign months.
The dashboard and the other tools we use for the campaign have been marked for several improvements by the community. We are currently working to evolve the functions of these tools to meet the dynamic needs of organizers and participants.
A campaign strategy for 2020 has to be developed with the Global team to ensure well defined roles and modes of engagement to improve the campaign’s growth.