Grants:Project/Eurecat/Community Health Metrics: Understanding Editor Drop-off
The primary value of Wikipedia is the editors. When an editor leaves the project, we lose their participation and contribution to the community. This could be related to multiple factors, also external to the project, but it could signal an issue related to internal dynamics and to the health of the community. While a big effort was dedicated to retain new editors, we lack knowledge and initiatives focused on understanding and preventing drop-off for experienced editors.
To identify factors associated with editors leaving the project, we plan to carry out an extensive study of the editor’s lifecycle. Special attention will be devoted to underrepresented groups according to social dimensions such as gender or geographic provenance. We will extend state of the art metrics to analyze different language editions, combining a computational approach with qualitative inspection of the findings, involving expert editors from the communities for this task when necessary. This will increase our understanding of factors that are important for community health in Wikipedia, and it will result in explainable metrics that can be applied to signal early if a page or set of pages are undergoing detrimental dynamics. We will make sure that our results can be translated into actionable knowledge to improve community health, and that our metrics and indicators will become available assets for the community to be acted upon. To this end, we will develop prototype dashboards that can be deployed in any Wikipedia language edition. To preserve privacy of the editors, only data aggregated at community level and page level will be shown.
What is the problem you're trying to solve?Edit
Editors make Wikipedia valuable. When an editor leaves the project, we lose their participation and contribution to the community. Multiple factors can influence the decision of leaving Wikipedia, also external to the project, but it could signal an issue related to internal dynamics and the health of the community. However, there is a limited understanding of the dynamics that affect the editors' lifecycle. We lack basic knowledge about the process: at which point in time of their Wikipedian life do people abandon the project? How many veteran editors are leaving the project today compared to the past? Basic statistics currently available, such as the number of active editors, do not capture the various facets of the editors' lifecycle. We also lack solid knowledge about the factors associated with editors leaving the project.
This issue is particularly relevant for underrepresented groups of editors: we know there is a gender gap in the editor community, so it would be especially helpful to understand when and why women editors leave the project; the same applies to other groups that lack representation on Wikipedia, according to aspects such as geographic provenance.
Conflict may affect editor participation, and especially for women editors. Having one's edits reverted reduces the probability of an editor to keep active; this is especially important for newcomers, and a significant effort was devoted to understand and mitigate the barriers for new editors, regulating the behavior of automated bots and promoting initiatives to welcome newcomers and guide them within the project.
While a big effort was dedicated to retaining new editors, we lack knowledge and initiatives focused on experienced editors. We argue that, for maintaining community health, knowledge of when and why editors leave the project should be available to as many language communities as possible, so that we can understand what is happening and decide how to act to prevent detrimental dynamics and get safer and more constructive spaces in the project.
What is your solution to this problem?Edit
To increase our knowledge of the dynamics of editor drop-off, we propose to run an extensive analysis of the edit history of multiple Wikipedia language editions. We will characterize the editors’ lifecycle and produce statistics to get a clearer picture of the phenomenon. To identify factors associated with drop-off, we will build on our previous work characterizing interactions in Wikipedia, and quantify different aspects of editor experience over time and along editors' lifecycle. We will devote special attention to underrepresented groups of users, according to aspects such as gender, country, and native language. To preserve the editors’ privacy, we will generate metrics and data aggregated at the page level. We will develop dashboards to make metrics and indicators available to the community to act upon the reported factors causing the drop-off.
- Community in the loop
We plan to work in strict contact with the community, building on lessons learned from previous experiences. To make sure that the metrics, analyses, and tools we develop will reflect the actual views and needs of the editors, we will adopt an iterative approach, involving the communities to get input and feedback at each step of the project. For the in-depth analysis of factors associated with drop-off, we will focus on the languages in which we are fluent and have closer contact with the communities. We will reach out to other communities to assess whether our findings apply to other language editions.
- Privacy considerations
We are aware that aggregating data and making them available could expose sensitive or personal information about editors. For this reason, we will make sure that the metrics and tools we make available are anonymized and aggregated at a community level or centered on pages and not on editors.
The project proposes the following five goals:
- Assess editor drop-off across multiple languages
- Get an overview of drop-off statistics in multiple language editions, by analyzing their edit history.
- Increase our understanding of the factors associated with editor drop-off
- Find evidence of interaction patterns that tend to be associated with editor drop-off. We will collect data about different kinds of interaction over time and study their relationship with editor drop-off.
- Understand drop-off dynamics for underrepresented groups
- Put a special focus on the factors that may affect participation for groups of editors that are under-represented in the project (e.g. by gender, territory, native language).
- Identify spaces potentially affected by detrimental interactions
- Help the community to detect pages or groups of pages that are undergoing a critical situation, that might negatively affect editors. We will develop metrics and indicators to detect such situations that may need attention, and eventually the intervention of the community.
- Increase awareness about community health
- Help the community to make sense of data by exposing page-level metrics and indicators developed into interactive dashboards. Disseminate the results using community channels and in the scientific community.
Fit with strategyEdit
We believe this proposal is well aligned with Wikimedia 2030 strategy. In particular, the scope of this project fits within three recommendations:
- Promote Sustainability and Resilience (Recommendation n. 1)
This recommendation states: "we must support people: a dynamic and often changing volunteer base able to bring new ideas, leaders, and methods for inclusion, and staff that provides the support systems and aid communities in building long-term partnerships to allow our outreach to expand." It aims at making communities more sustainable, considering that people is the organizational resource of the project.
Among different initiatives, the recommendation proposes to elaborate a list of indicators to ensure cultural change happens and measure them on a regular basis. We believe that, with the tools we propose to build, we can monitor community health and see when a specific community is more or less likely to suffer from editor drop off.
- Create Cultural Change for Inclusive Communities (Recommendation n. 2)
This recommendation states: "we need to make cultural change founded in Movement-wide standards for an inclusive, welcoming, safe, and collaborative environment that enables sustainability and future growth through extensive consultation with the involved communities.". It proposes to create a specific Code Of Conduct to establish basic community responsibilities and safeguarding and maintaining a healthy working atmosphere, and mechanisms to evaluate and test the effectiveness of policies.
We believe that with the tools we will build we can help the communities to monitor their health and assess the effectiveness of such policies.
- Improve User Experience (Recommendation n. 3)
This recommendation states: "we need to continually improve the design of our platforms to enable everyone—whatever their gender, culture, technological background, or physical and mental abilities — to enjoy a fluid, effective, and positive experience during both the consultation and contribution to knowledge throughout the Wikimedia ecosystem." It proposes many outcomes regarding the technology design and people's behavior in order to improve the environment in which newcomers and advanced editors contribute.
We believe that with the tools and research we will deploy we can "encourage communities to work towards shaping the projects to match the needs of those who are using them as knowledge consumers, along with those who are not yet participating as contributors". We want our community to be able to explore and understand more deeply the dynamics of interactions between members, in a way that could be translated, after community consultation, into better procedures and norms that reduce the attrition of users in the projects.
- Evaluate, Iterate, and Adapt (Recommendation n. 12)
This recommendation states: "we need to continually evaluate our progress toward our internal and external goals to be able to dynamically iterate, adapt, and upgrade our socio-technical processes and structures". It proposes outcomes such as "monitor progress on implementing equity in the Movement by evaluating diversity and newcomer inclusion as fundamentals of participation in the Movement", and "Provide evaluation with resources and experts on any given area."
We believe this project expresses fully the Evaluate, Iterate, and Adapt recommendation. With our metrics we will allow for monitoring spaces, community and governance, and having new tools to «increase self-awareness and accountability among all stakeholders, allowing us to adapt our strategies accordingly.»
How will you know if you have met your goals?Edit
In this section, we present the Outputs (what will be done by the end of the project) and the Outcomes (what will result from the project).
In parentheses, the goals associated with each output.
- A toolkit to compute editors' lifecycle and interaction metrics, complete with free and open-source code and documentation to ensure that our methods are fully reusable and reproducible (Goals 1-5);
- Editor lifecycle and drop-off statistics over the complete edit history for 30 Wikipedia language editions (Goal 1);
- In-depth analysis of factors affecting editor participation for at least 4 language editions: Catalan, English, Italian, Spanish (Goals 2-4);
- Feedback gathered from the community (Goals 2-4):
- at least 8 expert editors interviewed, 2 for each of the 4 languages above;
- at least 8 stories of editor drop-off, 2 for each of the 4 languages above;
- at least 8 expert editors from underrepresented groups interviewed;
- Prototype dashboards for 30 language editions (Goal 5);
- 1 video explaining how to use the dashboards;
- 1 scientific publication submitted to a high-level conference or journal (Goal 5);
- The community uses our dashboards to learn more about its state and health, and to take action:
- 1,000 pageviews within 6 months from the end of the project;
- 100 survey answers on the usefulness of the dashboards;
- usage stories from at least 10 language communities with lessons learned using our metrics;
- Community leaders use our results and tools to monitor the impact of their initiatives;
- Our code and metrics are used to develop further research and tools for the community;
- Our findings inspire new initiatives to prevent editor drop-off and improve community health.
First, we will compute editor lifecycle and drop-off statistics (Activity 1) for a set of 30 languages, the language editions with more than 1000 active editors. This will imply coming to a shared operational definition of drop-off, based on previous literature and feedback from the community, and will result in data that will be made public at the anonymized aggregated level.
Then we will run an in-depth analysis of editors’ interactions (Activity 2) on four selected language editions - Catalan, English, Italian, and Spanish - to find patterns associated with editor drop-off. We are limiting our analysis to these languages both for constraints of time and effort, and because these are languages that we understand and speak and for which we know the community, so we can qualitatively inspect the results and double-check whether what we are measuring corresponds to the actual interactions in the projects. We will rely mainly on language-independent metrics that we can apply to multiple languages. We will build on state of the art and develop metrics that can capture different aspects of editor interactions over time. As a non-exhaustive list, we will consider: edits in different namespaces; reverts, mutual reverts, revert chains; reverts with no personalized edit summary (empty, or auto-generated); having content deleted by another user; comments written, replies received, comments received in personal talk page, mutual replies, reply chains, discussion thread depth. Metrics will also account for the kind of editors involved in the interactions (anonymous, bots, registered, admin) and for their experience as measured by edit count. Language-dependent metrics will complement the analysis to quantify emotion and language features from messages written and received by editors, such as edit summaries from edits done, and from reverts done and received, comments written and received in talk pages, messages received in personal talk page and written on other users’ talk pages. We will also consider events affecting a specific page, such as when an editor is banned, or when a page is proposed for deletion, protected or semi-protected, or a template is added to a page (e.g., disputed content, non-neutral point of view). We plan to rely on the Mediawiki History Dumps as a base for mining the edit histories and compute interaction metrics, and on the XML dumps to complement these data when needed. We also plan to take advantage of existing tools and datasets with preprocessed data when available, such as WikiConv and WikiWho for the English Wikipedia.
These metrics will allow us to characterize the editor’s lifecycle, and look for dynamics associated with editor drop-off, through explainable models such as linear regression or random forests. We will validate our findings through qualitative inspection (Activity 3) with the support of expert editors.
To understand drop-off for underrepresented groups (Activity 4), we will first identify editors who publicly disclosed offline characteristics (gender, native language, or country of origin) through explicit means, such as userboxes and look at differences according to these social dimensions.
Once we have identified interaction patterns associated with editor drop-off, we will involve the community to select relevant metrics useful to nurture community health, and we will develop visual interfaces to expose them at the level of pages and sets of pages, or anonymized and aggregated at the community level. We will build a prototype dashboard allowing for interactive data exploration (Activity 5) for 30 language editions that can be expanded to a real-time system.
We aim to involve the community at every step of the process, so each activity should be regarded as iterative. We want the results of this study to be known in the community (Activity 6) and useful to support decisions. We will disseminate our findings also in academic conferences - publications will be open access and the code will be published with an open-source license - and we commit to make our research reusable and reproducible, within the privacy constraints related to editors' interactions.
- Work plan
- Compute editor lifecycle and drop-off statistics
- Identify existing definitions of drop-off from previous literature
- Elaborate and assess an operational definition of editor drop-off
- Compute and compare editor lifecycle and drop-off statistics on multiple language editions
- Define and compute longitudinal metrics of editor activity and interactions
- Compute language-independent metrics, for different Wikipedia editions
- Compute language-dependent metrics, for the English Wikipedia and other editions when relevant tools are available
- Identify factors associated with drop-off
- Apply mathematical models to characterize editor lifecycle and to detect patterns associated with drop-off, by editor and by page
- Compare results by language edition
- Qualitatively inspect the results
- Interview expert editors from the communities
- Analyse underrepresented groups of editors
- Identify groups of editors according to offline characteristics such as gender, country or native language, based on information disclosed by editors through userboxes, templates or other explicit means
- Identify factors associated with drop-off for different groups of editors
- Design and develop dashboards to increase community self-awareness
- Select relevant metrics and indicators
- Define appropriate visual models
- Test metrics and visual models and get feedback from the communities
- Implement dashboards in different languages
- Disseminate results
- Prepare and administer a survey about the usefulness of the dashboards
- Disseminate results in community channels and community events
- Disseminate scientific results
|1. Compute editor lifecycle and drop-off statistics|
|2. Define and compute longitudinal metrics of editor activity and interactions|
|3. Identify factors associated with drop-off|
|4. Analyse specific groups of editors|
|5. Design and develop dashboards to increase community self-awareness|
|6. Disseminate results|
- Total budget requested: 83,400 €
- Total personnel cost: 77,400 € (18 person-months × 4,300.00 €/person-month)
- Travel expenses: 6,000 € (including Wikimania, Wikimedia hackathon, one academic conference, and other relevant events)
- Project duration: 12 months
|1. Compute editor lifecycle and drop-off statistics||3|
|2. Define and compute longitudinal metrics of editor activity and interactions||3|
|3. Identify factors associated with drop-off||4|
|4. Analyse specific groups of editors||3|
|5. Design and develop dashboards to increase community self-awareness||3|
|6. Disseminate results||2|
Activities costs are calculated based at the following rate: 1 person-month = 4,300.00 €.
This corresponds to the rate applied by Eurecat for European projects and it does not include overhead/indirect costs, that will be covered by Eurecat.
Community engagement is a crucial activity to disseminate the value of the conclusions derived from the data, and to transform them into positive action across the entire Wikimedia movement. We plan to reach these projects:
- User groups (we list only the groups for which we already have a contact)
- Off-line events. The project will be presented in the following events:
Following the communications received from Wikimedia Foundation's grant officers, we are adding this section to the project proposal.
General focus of the projectEdit
Travel and offline activities and events constitute a minor focus of the project. The core activities of the project can be performed while teleworking, and all the core activities of the project will be done online and remotely. All the members of the team are currently teleworking from home since mid-March 2020.
One member of the team was planning to participate in the Wikimedia Hackathon 2020 in Tirana, Albania, which has been canceled. However, no funding for this activity was included in this proposal. It may be possible that the event will move online, in which case we will participate remotely.
All other offline activities related to the project were part of the following tasks (see #Activities):
- 6.2 Disseminate results in community channels and community events
- 6.3 Disseminate scientific results
These activities were planned for 2021 (Wikimedia Hackathon 2021, Wikimania 2021, and academic conferences). All these activities are scheduled well beyond September 15, 2020, the current date set by Wikimedia Foundation for blocking all offline events. We hope these activities will take place as planned.
In order to design, plan and execute the goals of the project we count on a group of participants and advisors with a diverse set of skills and expertise on the topic.
- CristianCantoro as grantee/researcher. Cristian Consonni, Ph.D., is a research scientist in computational social science at the Data Science & Big Data Analytics unit in Eurecat, Centre Tecnològic de Catalunya. Cristian obtained his Ph.D. in Computer Science from the University of Trento with a thesis on the emerging knowledge from the structure of links in Wikipedia. He has been editing Wikipedia since 2007 and he has been active in the broader Wikimedia community since then: he has served on the board of Wikimedia Italia from 2007 to 2017 in various roles; he has been a member of the international organizing committee of Wiki Loves Monuments 2013; and from June 2013 to June 2015, he has been a member of the Wikimedia Foundation's Funds Dissemination Committee.
- Sdivad as grantee/researcher. David Laniado, Ph.D., is a senior research scientist at the Data Science & Big Data Analytics unit in Eurecat, Centre Tecnològic de Catalunya. Sdivad has extensive experience in the study of online collaboration, in particular, he has published over 20 academic papers on different aspects of social interactions in Wikipedia. He is a co-creator of the Contropedia platform for the analysis and visualization of controversies in Wikipedia articles.
- Elaragon as grantee/researcher. Pablo Aragon, Ph.D., is a research scientist at the Data Science & Big Data Analytics unit in Eurecat, Centre Tecnològic de Catalunya. Pablo's research focuses on understanding social and political phenomena through the analysis of data from the Internet. Pablo is a board member of Decidim, the platform for participatory democracy launched by Barcelona City Council and co-founder Democratic Innovation Lab of Barcelona. He also served as a data scientist in ParticipaLab - Medialab Prado for Decide Madrid, the platform for direct democracy launched by Madrid City Council.
- MNovotny (WMF) as advisor. Margeigh Novotny is a Senior Director of Product Design Strategy in the Audiences Team at the Wikimedia Foundation, leading product strategy, and design strategy efforts. Her focus is on looking ahead of where "the product" is currently, to understand how the user experience needs to adapt and evolve as we move toward our 2030 goal of providing all the world's knowledge to all the world's people, for free.
- Miriam (WMF) as advisor. Miriam Redi is a Research Scientist at the Wikimedia Foundation and Visiting Research Fellow at King's College London. Formerly, she worked as a Research Scientist at Yahoo Labs in Barcelona and London, and Nokia Bell Labs in Cambridge. She received her Ph.D. from EURECOM, Sophia Antipolis. She conducts research in social multimedia computing, working on fair, interpretable, multimodal machine learning solutions to improve knowledge equity.
- PatriHorrillo as advisor. Patricia Horrillo is an expert journalist in communication and social networks. Since 2015, she coordinates the Wikiesfera UG  at Medialab-Prado (Madrid, Spain), a weekly working group where she helps more people to learn how to edit in Wikipedia and to know better other Wikimedia projects. She is concerned about the gender gap and organizes thematic edithatons to make more relevant women visible in Wikipedia.
- Halfak (WMF) as advisor. Dr. Halfaker is a principal research scientist at the Wikimedia Foundation and a senior scientist at the University of Minnesota. He studies the intersection of advanced algorithmic technologies and social issues in open production communities (like Wikipedia) using a mixture of experimental engineering, data science, and ethnographic methods. He's most notable for his studies of Wikipedia’s editor decline and his development of “ORES”, an open AI platform for Wikipedians.
- marcmiquel as advisor. Marc Miquel-Ribé, Ph.D., holds a doctorate from Universitat Pompeu Fabra in Barcelona with a thesis centered on Wikipedia and the study of the Editor Engagement, Cultural diversity, and identities in different Wikipedia language versions. He is a member of Amical Wikimedia (Catalan Wikipedia) since 2011. He has given numerous talks on cultural diversity, user experience and editor engagement in Wikimedia events. Recently, he has been leading the research work of the Wikipedia Cultural Diversity Observatory. Additionally, he's been one of the writers of the Wikimedia Strategy 2030 Plan and helped shape the narrative of the Cultural Change and was involved in the final writing and refining of the document.
- Eurecat - Centre Tecnòlogic de Catalunya is a non-profit technology center created in 2015, through the merging of the most important technology and research centers in Catalonia. Eurecat is the second-largest private research organization in Southern Europe, with more than 650 professionals participating in more than 200 research and development projects. Eurecat is currently participating in more than 60 EU-funded projects, mainly in the Horizon 2020 Programme.
We plan to notify the communities on their specific communication spaces as well as to give detailed information in the main Wikimedia communication channels.
- Mailing lists:
- Strategy Working groups (via mailing lists):
We are aware of previous and current efforts in understanding community health. We are in touch with stakeholders all over the Movement in order to both learn from their projects and data as well as to share ours. These are some of these projects:
- Metrics kit, part of the Community health initiative, a project developed by the Trust and Safety team at WMF. The Metrics kit project seems to be currently frozen since 2019 due to "a lack of funding".
- Editor Interaction Data Extraction and Visualization an Individual Engagement Grant proposal over a similar topic. The project was focused on studying editor interactions in general. In this project, we are interested in studying factors associated with editor drop-off.
- "Measuring community health - Vital Signs for Wikimedia projects" a presentation by Dario Taraborelli, Aaron Halfaker, and Dan Andreescu from Wikimania 2014. This work led to the development of the metrics tool - later renamed event metrics - and it is more focused around new editors and editors participating in events. Related to that is the outreach dashboard.
- Lam, S. T. K., Uduwage, A., Dong, Z., Sen, S., Musicant, D. R., Terveen, L., & Riedl, J. (2011). WP: clubhouse? An exploration of Wikipedia's gender imbalance. In Proceedings of the 7th international symposium on Wikis and open collaboration.
- Hill, B. M., & Shaw, A. (2013). The Wikipedia gender gap revisited: Characterizing survey response bias with propensity score estimation. PloS one, 8(6).
- Laniado, D., Kaltenbrunner, A., Castillo, C., & Morell, M. F. (2012). Emotions and dialogue in a peer-production community: the case of Wikipedia. In Proceedings of the eighth annual international symposium on Wikis and open collaboration (pp. 1-10).
- Bear, J. B., & Collier, B. (2016). Where are the women in Wikipedia? Understanding the different psychological experiences of men and women in Wikipedia. Sex Roles, 74(5-6), 254-265.
- Halfaker, A., Kittur, A., & Riedl, J. (2011). Don't bite the newbies: how reverts affect the quantity and quality of Wikipedia work. In Proceedings of the 7th international symposium on wikis and open collaboration (pp. 163-172).
- Smith, C., Yu, B., Srivastava, A., Halfaker, A., Terveen, L., & Zhu, H. (2020). Keeping Community in the Loop: Understanding Wikipedia Stakeholder Values for Machine Learning-Based Systems. arXiv preprint arXiv:2001.04879.
- Panciera, K., Halfaker, A., & Terveen, L. (2009). Wikipedians are born, not made: a study of power editors on Wikipedia. In Proceedings of the ACM 2009 international conference on Supporting group work.
- Algan, Y., Benkler, Y., Fuster Morell, M., & Hergueux, J. (2013). Cooperation in a Peer Production Economy Experimental Evidence from Wikipedia. Available at SSRN 2843518.
- Flöck, F., Erdogan, K., & Acosta, M. (2017, May). TokTrack: a complete token provenance and change tracking dataset for the English Wikipedia. In Eleventh International AAAI Conference on Web and Social Media.
- Laniado, D., Tasso, R., Volkovich, Y., & Kaltenbrunner, A. (2011). When the Wikipedians talk: Network and tree structure of Wikipedia discussion pages. In Fifth International AAAI Conference on Weblogs and Social Media.
- Kaltenbrunner, A., & Laniado, D. (2012). There is no deadline: time evolution of Wikipedia discussions. In Proceedings of the Eighth Annual International Symposium on Wikis and Open Collaboration (pp. 1-10).
- Yasseri, T., Spoerri, A., Graham, M., & Kertész, J. (2014). The most controversial topics in Wikipedia. Global Wikipedia: International and cross-cultural issues in online collaboration, 25.
- Iosub, D., Laniado, D., Castillo, C., Morell, M. F., & Kaltenbrunner, A. (2014). Emotions under discussion: Gender, status and communication in online collaboration. PloS one, 9(8).
- Neff, J. J., Laniado, D., Kappler, K. E., Volkovich, Y., Aragón, P., & Kaltenbrunner, A. (2013). Jointly they edit: Examining the impact of community identification on political interaction in Wikipedia. PloS one, 8(4).
- As per a message from Ryan Kaldari on Analytics mailing list on Thursday, February 27.
Do you think this project should be selected for a Project Grant? Please add your name and rationale for endorsing this project below! (Other constructive feedback is welcome on the discussion page).
- Support That's an important research topic, and Cristian and the other participants will be able to conduct it. Jaqen (talk) 18:06, 21 February 2020 (UTC)
- Support This project seems to be one of the fittest for getting data about user drop-off/retention. It is also good to understand the dynamics about some of the older communities, in order to fix the issues they're having and try to prevent those issues to repeat in younger, other-than-European communities. Sannita - not just another it.wiki sysop 12:05, 23 February 2020 (UTC)
- Support It’s an important thing to focus on.--Ferdi2005[Mail] 17:27, 24 February 2020 (UTC)
- Support I'm in favour Girardelli Simone (talk) 19:06, 24 February 2020 (UTC)
- Support a timely and needed study. Pundit (talk) 18:07, 24 February 2020 (UTC)
- Solid plan and advisory board collected for this complex and important topic. Ainali (talk) 20:16, 24 February 2020 (UTC)
- Support Important topic which requires good research. MuRe (talk) 21:07, 24 February 2020 (UTC)
- Support it's important to evaluate issues to arrive at solutions. SusunW (talk) 21:09, 24 February 2020 (UTC)
- Support This seems like a very well-defined and well-thought-out proposal, and the topic is of course very important. Jon Harald Søby (WMNO) (talk) 10:40, 25 February 2020 (UTC)
- Support This project could give us valuable understanding of what causes editor drop off, which is particularly relevant for underrepresented groups of editors, and the proposal is aligned with the 2030 Wikimedia Strategy recommendations. Astrid Carlsen (WMNO) (talk) 10:48, 25 February 2020 (UTC)
- Support I think it is valuable to understand when editors drop-off, and at what stage of their 'Wikilife'. This project could also give some insight to why some people choose to leave Wikimedia projects and others don't. Netha Hussain (talk) 17:20, 25 February 2020 (UTC)
- Support Any organization needs to know its resources, and Wikipedia is no exception. Filippof (talk) 19:19, 25 February 2020 (UTC)
- Support Very important topic to focus on these days. Good luck with the research and looking forward to seeing the results. - CEllen (talk) 19:27, 25 February 2020 (UTC)
- Strong support. Editor retention or new editor retention depends on how receptive the editor community is. If there is no proper protection in place, people will simply give up, get depressed and leave like what is being schooled here in this talk page. --Exec8 (talk) 11:07, 26 February 2020 (UTC)
- Strong support. Is about community health and users retention, is about diversity and unrepresented groups, is about metrics, software and data, that everyone can benefit. And, as people matters, my plain trust and respect for the team of applicants and advisors. Camelia (talk) 11:24, 26 February 2020 (UTC)
- Strong support valuable topic & the team will be able to deal with it --Vale93b (talk) 12:08, 26 February 2020 (UTC)
- Strong support for important research. --Rosiestep (talk) 16:59, 26 February 2020 (UTC)
- Support It is a good research area, and in fact mot much information is available about it. I have a number of observations that I am noting in the discussion page, but overall, I endorse the proposal and I believe that this work should really happen! -- Anass Sedrati (talk) 17:51, 26 February 2020 (UTC)
- Support. Information from this kind of study is necessary for informed planning. · · · Peter (Southwood) (talk): 19:23, 26 February 2020 (UTC)
- Support - thank you all to the people involve for initiating this. I totally support this very well written grant proposal. --Greta
- Support - great idea! Chris Keating (The Land) (talk) 20:10, 27 February 2020 (UTC)
- Vital to the mission. Benjamin (talk) 08:49, 28 February 2020 (UTC)
- Support The project is important for the community, and I belive on it.--Reda Kerbouche (talk) 17:28, 28 February 2020 (UTC)
- Support The project is important for the community. Dgw (talk) 18:17, 28 February 2020 (UTC)
- Support I have been asked to provide feedback on this project proposal by a member of the project team, online, in a direct contact. I find the relevance of this research project to be unquestionable and fully endorse it. I have placed my comments on the Talk Page, section Endorsement and Suggestions. GoranSM (talk) 08:12, 29 February 2020 (UTC)
- Support Looks like an interesting project. --Fjmustak (talk) 18:45, 29 February 2020 (UTC)
- Support Well thought of research project proporsal, I think it needs to be given a chance and support. I'm keen to see the outcomes. Bobbyshabangu (talk) 12:08, 3 March 2020 (UTC)
- Support I think this project can improve knowledge of how editors become involved in Wikipedia projects or change how they participate. It seems like this work has a decent empirical base to build upon, but that mostly derives from English Wikipedia. Studying other language editions can help us explore variation between projects. The researchers might also consider studying editors of non-wikimedia wikis. Nemo recently completed an archive of the Fandom wikifarm which has many thousands of active Wikis. I also like the idea of building tools or dashboards to help communities learn about themselves using the metrics developed in this research. Groceryheist (talk) 17:06, 3 March 2020 (UTC)
- Support Really interesting project about community health. We need it. Marcok (talk) 19:05, 3 March 2020 (UTC)
- Support Interesting topic, that can increase Wikimedia projects "lifestyle", and decrease the amount of users that drops off ValeJappo (talk) 16:56, 4 March 2020 (UTC)
- Oppose I'm sorry, but to me this seems more of an academic research, which will have little practical impact on the community. I have also some doubts that it is possible to truly give a reasonable picture of the drop off statistics using this essentially purely metric approach. Finally, the timeline seems over-stretched to me considering the project will have 1.5 persons working full time on it. Sorry. --Sandrobt (talk) 07:26, 5 March 2020 (UTC)
- Support I think that editor retention is important, and that having a more detailed understanding about the factors involved in editors leaving Wikipedia is valuable. I think this has a practical impact on the community because editor retention affects the health of our community. The research proposal seems well-thought out, but I must admit that I don't have much experience with reading research proposals to have any other relevant comparisons. Clovermoss (talk) 04:19, 8 March 2020 (UTC)
- Support I believe in the need and importance to study and understand the Community Health Metrics. It can help in lot many ways and I totally support it. Best Wishes-Manavpreet Kaur (talk) 15:22, 11 March 2020 (UTC)
- I endorse this project and the team working on it. But... I must add that I have been wondering why it would be proposed as a community project grant rather than going through a WMF department (they have budget for those probably ?). Anthere (talk)
- Effeietsanders (talk) 01:18, 13 March 2020 (UTC) Sounds excellent.
- Support I think this project research will give some insight understandings which are important for community health. My best wishes-Rajeeb (talk!) 13:55, 15 March 2020 (UTC)
- Strong support We need to learn more about user retention and how it affects community health. Trust and respect for the team. Tiputini (talk) 15:50, 15 March 2020 (UTC)
- Strong support A highly commendable project worthy of consideration. It will avail us with community health data which will help make future informed decisions. Editors are the heart of this community. And knowing how to preserve the fragility and health of this "heart" is vital to community growth and development. This project will walk the community in that direction. I'm eager to see the outcome of this research project. Ptinphusmia(talk) 12:13, 16 March 2020 (UTC)
- Support Great idea. Even if as someone said before "It's academic", I think it's needed to understand the community. Hiperterminal (talk) 10:34, 17 March 2020 (UTC)
- Strong support I know I'm appallingly late, but hopefully I have something useful to add. The advisors, and especially @Marcmiquel, know what we can produce from the Hadoop cluster to support research like this. I think this project deserves the best chance of success we can give it, so I'll do everything I can to help you gather numbers and present them. Please do catch me on IRC or facebook or wherever and we can talk specifics. Milimetric (WMF) (talk)