Things that extend a wiki career

There is a natural tendency in this sort of study to focus on the things that drive people away, but there are also things that motivate people to stay here. I would like to see some stats as to whether any of the following are true or false:

1. Gaining an awarded user right such as rollbacker or reviewer extends ones wiki career. In the last year we have appointed many hundreds of Autopatrollers on EN Wiki as well as large numbers of reviewers. Analogies with other online communities imply that this is likely to encourage people to stay on, but it would be nice to know.
2. Does passing RFA or RFB and becoming an admin or crat extend a wiki career and motivate people to stay longer?
3. Conversely does losing a useright or an RFA/RFB drive a proportion of editors away?
4. Does receiving barnstars motivate people to stay longer? Or are the sort of editors who do things worthy of barnstars the sort of editors who are motivated to stay anyway?
5. Does physical geography and the location of our servers have anything to do with retention? Do the Amsterdam servers give us better retention rates amongst European editors and would a similar facility in India or Brazil boost our editing communities there? East Africa should be a good test for this, recently there was a major upgrade in their connection to the Internet and it would be interesting to know whether this meant:

1. The same editors did the same sort of edits but it took them less time.
2. The same editors spent the same time on wikipedia and did more edits because the connection was faster.
3. We gained more editors.

1. Does attendance at real life wiki events such as meetups, manias and Glam/Wiki editing collaborations tend to result in better retention of editors.

Hi WereSpielChequers,

You are raising a whole number of very valid questions! From a methodological point they are challenging because we can never say that someone left the community permanently, maybe they are on an infinite Wikibreak. So the challenge will be to come up with a measurement that accurately describes the lack of engagement for a prolonged period of time. Do you have any ideas on how to measure this?

Best, Diederik

Hi Diederik, I think a useful approach on this would be to analyse wiki breaks. It would be helpful to know of the people who have stopped editing for three months how many edit again after 1, 2, 3 or more months. Once we know that only y% of editors who have been inactive for 36 months are likely to return in a given month then we can be more robust about predicting how many future inactives will return. Of course the project is far too young to know how many will return after gaps of a decade or more. But I would anticipate that if we charted length of wikibreaks we would see a pattern and could start extrapolating and making predictions as to future returning Wikipedianss. At some point in the future we will probably get an idea of:

1. How long a break needs to be before we can be 99% confident that the editor will not return.
2. Whether editors of the "founder generation" (2001-2005) will have a different pattern here to later generations.
3. Whether editors have a different reactivation pattern if they are active on other Wikimedia projects - currently we analyse each project quite separately.
4. Whether the chance of someone returning is linked to their number of edits or usergroups
5. How blocks, bans, exercising right to vanish or posting a retirement template alter your chance of returning.

Also I would dispute that we can never say an account is permanently closed. We know that some former editors are dead, we don't currently have an appropriate distinct flag for that though we probably need one (I would predict that by 2100 most of our current editors will be dead). We also know that some people have been caught abusing multiple accounts and even if they are unblocked usually all but one of those accounts will be permanently blocked.

Hi WereSpielChequers,

We do consider wiki breaks in the current Editor Trends Study when calculating the tenure of a Wikimedia editor but I think you raise some interesting additional questions. I think we should collect these additional questions (maybe on the page Editor_Trends_Study/Future_Research and see how we can answer these.

Best, Diederik

I found a study that shows editors who participate in talk page discussions are more likely to stay. Same with editors who participate in improving FAs. Editors who participate in both are MUCH more likely to stay.

Of course, this could be a correlation and not a cause. Someone who is engaged with improving featured articles and collaborating is the kind of person who would already find Wikipedia more interesting. But it would be interesting to see if it would help to nudge someone towards a good WikiProject that's very collaborative and interested in featured articles. I know it stepped up my participation a lot.

Randomran, can you post a link to the study you're referring to? Thanks!

There's a lot of good stuff in the community health recommendations. Didn't take long to dig it up. Hope it's helpful! Click on the image and it will take you to a quote I pulled from the study. It's a great, great study.

Re: 1. The editors have to ask for those privileges. I'd assume they expect them to get them, and that getting them is not a big deal; however, not getting them could be, and could lead to them leaving the community. Re: 2. Ditto, plus an anegdotical evidence that I know at least one editor who, as one of the reasons for leaving the project, mentioned that his 2nd RfA was "shut down" by his "enemies". Re: 3. I wouldn't be surprised, although this is a very rare occurrence. Re: 4. I'd think it does. See my thoughts on that in a mini-essay. Re: 5. I doubt it, speed should be the same, worldwide, controlling for end-user infrastructure. Re. 6. I'd expect it does. This needs more studies, certainly. --Piotrus 23:26, 4 March 2011 (UTC)

Hi Piotrus,

Editors don't always have to ask for privileges. We've done some interesting work on EN wiki creating prospect lists of people who should have the Reviewer Autopatroller or even I think the Rollback flag. My experience of turning up on people's talkpage, complementing their work and setting their account as an Autopatroller is that people take that very positively, and the ones I check who've recently had articles deleted or been warned about copyvio wouldn't know that I'd looked and moved on.

As for Internet speed, I'm pretty sure that it does vary around the world according not only to the speed of the local connection to the ISP but also connections further afield. Certainly my experience of editing in South America was that local sites might be OK but Wikipedia and also UK sites were much slower there than when I'm in London. My understanding is that the squid servers in Amsterdam are meant to increase speeds for editors here in Europe, and the new data centre in Virginia will speed things up by being closer to Internet hubs.