Research talk:Wikipedia Administrator Recruitment, Retention, and Attrition

Research Signup

If you're interested in following along this research project, or would like to be contacted as a participant for this research, please leave a reply below. —User:CLo (WMF) (talk) 19:22, 30 July 2024 (UTC)

Prompts on administrator motivations

As part of our scoping work, we'd like to ask you to answer one of these prompts! The replies to this will help us design our future interview and survey guides.

If you are more comfortable leaving comments in a language other than English, please feel welcome to do so. Please note that we may use machine translation in reviewing feedback provided in languages other than English.

• If you are a current administrator: what was your top reason for becoming an administrator?
• If you are not an administrator, but have ever considered becoming one: what has been your top reason for wanting to be an administrator?
• If you have previously been an administrator, but no longer are: what is (or was) your top reason for leaving the administrator role?
• If you have ever actively avoided becoming an administrator, what is (or was) your top reason for not wanting to be an administrator?

—User:CLo (WMF) (talk) 18:56, 1 August 2024 (UTC)

I was actively avoiding becoming an admin because this role on ukwiki requires a lot of conflict solving which I'm not comfortable at.

I am currently an admin and I became one, when someone finally persuaded me that I can stick to the other admin responsibilities I'm OK with. Besides, given the current situation in Ukraine I wanted our community to have a spare admin just in case. Ата ^(talk) 07:21, 9 August 2024 (UTC)

Hi @Ата, thanks for sharing about your experience, it's good to understand what your hesitations/concerns were, and what finally persuaded you to consider performing admin responsibilities. If you're comfortable sharing more (completely optional and up to you), I would be interested to know what specific admin responsibilities you generally prefer for your day-to-day work, as well as what types of examples of conflict solving that initially made you hesitant to become an admin. Now that you're an admin, have you successfully been able to avoid these conflict solving situations that you weren't interested in becoming involved in? Thanks again for sharing about your experience. EAsikingarmager (WMF) (talk) 16:23, 17 September 2024 (UTC)

EAsikingarmager (WMF), so far I have only deleted pages in obvious cases (unused templates, empty categories, test articles), restored maybe one article, blocked five users for inappropriate usernames, protected an article once, and I'll stick to these actions in the future. I have indeed successfully been able to avoid conflict solving situations, which mainly occur on Requests to admins page. Ата ^(talk) 15:05, 19 September 2024 (UTC)

• Thanks for inviting us to comment at Wikimania! I can answer (1) and (3) from my experience:
  • Why did I become an administrator? I was doing curatorial tasks for which getting the delete and protect buttons would be useful, was encouraged by fellow administrators, and was young and wanted to collect an extra hat (such behaviour is officially discouraged by policy but I admit to my former self being like that). Once you become an admin for one project, it's easier to lean on your own experience in deciding whether to ask for adminship on another project.
  • Why did I leave? I have given up some permissions voluntarily because I no longer had the time to help with those projects. I gave up the advanced permission as each unused advanced permission is theoretically a security vulnerability. Deryck C. (talk) 10:33, 9 August 2024 (UTC)

• i became an admin because it gave me access to protected pages which i frequently needed to edit, because of my work with gadgets and templates.
  • I stopped being an admin because i no longer felt supported by the community in the expectations that i had of senior/long term editors and their behavior, as well as enforcement of those expectations that we were setting for new contributors
  • —TheDJ (talk • contribs) 10:33, 10 August 2024 (UTC)

  Hi @TheDJ, thanks for sharing your experience. If you'd be interested and willing to share more (completely optional, of course), I'm curious what types and examples of support might have made you feel more supported? Thanks, EAsikingarmager (WMF) (talk) 16:29, 17 September 2024 (UTC)

I have never been an administrator, but considered becoming one: my top reason for wanting to be an administrator is to go on the next level of wiki proficiency. Sofiemama (talk) 11:22, 10 August 2024 (UTC)

Hi @Sofiemama, thanks for sharing about your top reason. If you're comfortable and open to sharing more, I'd like to understand more about what it means to you to "go to the next level of wiki proficiency". Could you share some more details about what that means to you, and why it's important. Are you still considering becoming an admin? If so, what, if anything, has prevented you from beginning the process. What would encourage you to begin the process? Thanks again for sharing, EAsikingarmager (WMF) (talk) 20:23, 18 September 2024 (UTC)

Data collected

User:Samwalton9 (WMF) noted that you're collecting data about requests for adminship on Discord. One data point I'd encourage you to collect how RFAs are publicized (at enwiki the watchlist notification is a turn off for many potential admin I've talked to) and (at least for enwiki, don't know about your other studied wikis) the number of questions asked of candidates (even uncontroversial RFAs often require a huge amount of time from candidates because of these questions). Best, Barkeep49 (talk) 15:52, 2 September 2024 (UTC)

Hey @Barkeep49! Yes, one of the things I've been doing for this project is looking at publicly-available data on RFAs across our wiki shortlist, and that of course includes English Wikipedia RFAs. Unfortunately, statistics about number of questions (or even number of comments) per candidate are not generally widely available on Wikipedias aside from English, so we don't have a great comparative basis for that. Nevertheless, I did look broadly at English Wikipedia RFAs, and something that did jump out to us was the high number of comments per RFA.

I collected information from the en:RFAs by year page, and calculated "comments" as the sum of S + O + N. I then categorized each RFA by outcome into two buckets, "successful" or "unsuccessful". Looking purely at the Jan 2018-Jul 2024 period, the mean number of comments per English Wikipedia RFA are as follows:

• Successful RFAs: 223.59
• Unsuccessful RFAs: 87.92
• All RFAs: 175.98

Now from those aggregates, I did not further break down success versus neutral versus objections, and of course these only encompass top-level comments and doesn't necessarily capture the full breadth or length of some of these discussions. But I think it's safe to say, even at a glance, that a successful RFA seems to generate or require an enormous commitment to dialogue with other contributors and functionaries. I would attribute the lower figure for unsuccessful RFAs to the fact that many unsuccessful RFAs are closed faster, due to WP:NOTNOW or WP:SNOW reasons. But an average of ~88 top-level comments per unsuccessful RFA still seems like a lot of time and energy invested into the attempt.

I should note that this jump in number of comments per RFA can be pinned down to a point around 2010/2011, coinciding with the introduction of administrator activity requirements, but I don't know enough about the history of RFAs to paint any causal relationship there. I'm currently working with my colleagues on a plan to release this data (and many more interesting preliminary metrics!) publicly, as part of this project.

Happy to discuss more on the subject! —User:CLo (WMF) (talk) 17:36, 17 September 2024 (UTC)

Claudia - you may be interested in work @Hey man im josh has done on questions. Best, Barkeep49 (talk) 18:57, 17 September 2024 (UTC)

Thanks for the tip! —User:CLo (WMF) (talk) 15:43, 19 September 2024 (UTC)

Retention strategy

I am involved in other, real-life leadership roles and I would say that Wikipedia is the only volunteer project which tolerates abusive users to that extent. But that's not news (and shouldn't be) to anyone who is deeply involved in the Wikimedia movement. I think the biggest retention issue is that people don't know how much abuse they will receive after becoming an admin. One part is dealing with on-wiki process like blocking users, dealing with appeals. You will encounter trolls, people threatening violence and occasionally stalkers/doxxers. Another part is dealing with off-wiki, often legal issues. Some examples include: imprisonment, threatened arrest, exerted government pressure on their editing and losing their employment. Essentially, admins are expected to do voluntary work on behalf of their wikis (and WMF in public's eyes) with absolutely no support network, token of appreciation or even a word of thanks on the talk page (unless something is extremely bad and had to get Trust & Safety or Legal involved).

In the past, WMF takes a passive role which allowed these behaviours to fester. And when these behaviours were left unchecked, others saw it as a community dynamics that is tolerated and they themselves would also do it (i.e. broken windows theory). It also made future admins unwilling to enforce on civility rules because they would be viewed as "overly harsh" admins with little precedence to rely on when a behavioral block is warranted. What WMF can do now is taking a more active role on handling and removing harassments. Since Foundation probably can't provide mental health service or assistance program to admins, the Foundation should consider forming support group for admins who encounter abusive behaviour so that there is a safe space and to identify any cross-wiki behavioral abuse. Finally, there should be some sort of admin recognition program to recognize their contributions to the project. OhanaUnited^{Talk page} 18:08, 12 September 2024 (UTC)

Hi @OhanaUnited, thanks for commenting and sharing about your experience. I appreciate that you've identified some specific actions you think would be helpful, including the idea of a support group for admins and admin recognition program. In terms of how you envision those two things, is there any additional detail you'd like to share? (for example, such as how you'd like to see the support groups work, or recognition ideally look like). Thanks, EAsikingarmager (WMF) (talk) 16:59, 17 September 2024 (UTC)

For support group, it could be a dedicated staff or point of contact that admins can informally reach out when admins want resources on dealing with persistent block appeals, "low level" paid editing, inserting hoax/falsify reference and a lot of the cases listed in long-term abuse (in English and Chinese for example), i.e. things that are not serious enough to escalate to T&S. The staff can provide links and resources on where and how to get help. The dedicated staff can also assist with identifying and coordinating a solution for cross-wiki behavioral issues, which are subjected to different community rules, such that the issue can be resolved while following their respective communities' rules. It just so happens that we have an ongoing example this week on a block from ja.wp that the discussion spreads over to en.wp. Volunteers are unlikely to spend time to read up the conversation or rules in another language, often machine translated and missing cultural nuances, and lead to votes & decisions that are not based on full pictures. And when decisions appear to be unjust due to mistranslation, this is how it breeds long-term abusers as they see the system to be unfriendly, unequal enforcement of rules and target admins for continuing to support the system.

Admin T-shirt (text to be edited)

Now that we covered the "push" factors, I'll talk about the "pull" factors. For recognition, the only time I see someone being thanked for their service is when they voluntarily hand in their admin tools because they're quitting or no longer wants to be an admin. I have been an admin on en.wp for 16 years and there was never a note left on my talk page to thank my service on any of my "admin birthday". If the budget is $0, at least have a robot to leave a talk page message on the "admin birthday" to thank them for X years of service and show something like w:Special:Impact for number of admin actions taken in the past year. If there is some budget for this initiative, then providing T-shirt or small lapel pins like these as part of the onboarding experience. There can be different variations of label pins for every 5 years of admin service. These merchandise also have the benefit of conversation starters with the public and at outreach events. The benefit of enhanced tracking can also identify admins who just recently became inactive and see whether it was temporary or permanent. OhanaUnited^{Talk page} 20:33, 17 September 2024 (UTC)