Research talk:Alternative lifecycles of new users

From Meta, a Wikimedia project coordination wiki

Other routes[edit]

Hi, some other routes worth thinking about are:

  1. The new article creators 24.7% of new editors on EN wiki in this recent study. Though the proportion of these who stay is very low.
  2. Vandal fighters, people who see an obvious vandalism on Wikipedia and fix it. One theory for the slowdown in recruitment is that less vandalism is now visible as the bots are reverting the vast majority of it the same moment it happens, and the downside of this is a drop in recruitment as fewer readers are seeing vandalism that provokes them into editing.
  3. Typo fixers. As with the vandalfighters the theory is that these are being crowded out by semi automated editing and other "gnomish" activity. Typos are still easily found amongst new pages and long ones written by non-native speakers (Firefox's spellchecker and possibly others only highlights possible typos for a certain number of bytes of an article - beyond that they are less likely to be fixed). But the casual reader is much less likely to come across a typo than they were in earlier years.
  4. Goodfaith inhabitants of assorted pitfalls. EN wiki supports multiple spellings/dialects of English, and allows both CE/BCE and AD/BC dating, providing you are consistent within articles and rarely if ever shift an article from one version of English to another. So a proportion of new editors start by "correcting" what they think is a typo or an anachronism and are liable to be reverted, often with a fairly terse comment. It would be interesting to know whether this is increasing in line with readership or if the pattern is more related to new readers among native speakers; Established readers may be used to the spelling variation and non-native speakers less confident about fixing possible typos. This is one area that will vary between wikis as Project policies are radically different. Dutch and Deutsch I believe each have a single standardised approved spelling for their projects, the Portuguese language wiki includes both Brazilian and Portuguese spellings.
  5. Updaters. These are people who add new match results, film roles, release dates and dates of death to existing articles. Such editors almost invariably start off adding unsourced information, I suspect in some cases they are telling us before they tell the papers. On the plus side the potential for such edits may actually be increasing in line with the number of articles, or of articles with a contemporary aspect. On the downside the increased trend towards requiring sources and the reverting of unsourced information means that these editors as a group are likely to be getting a chillier response than their predecessors WereSpielChequers 22:09, 6 June 2011 (UTC)Reply[reply]
Thanks for these categories, I think it'd be really interesting to develop a typology of new users, and see which ones are more or less likely to continue editing x months after they first edit. Staeiou 17:18, 13 June 2011 (UTC)Reply[reply]
This reminds me of some studies discussed at the 2010 Wikimania that used experience from the Red Cross and their different types of volunteer. WereSpielChequers 14:08, 6 July 2011 (UTC)Reply[reply]

Policy callers[edit]

I'd be interesting in seeing some exploration in the direction suggested by Beschastnikh et al.[1] who found that every week over 10% of editors citing policies are first-timers. Citing a policy as an early act is at odds with the standard view of a new editor's lifecycle. --DarTar 14:53, 7 June 2011 (UTC)Reply[reply]

I think this would fit best with Research:How do editors work anonymously? as in my experience one of the most common explanations is that such editors are longstanding IP editors who have finally decided to create a user account. WereSpielChequers 14:22, 8 June 2011 (UTC)Reply[reply]
  1. Beschastnikh, I., Kriplean, T., and Mcdonald, D. W. Wikipedian Self-Governance in action: Motivating the policy lens. In Proceedings of the 2008 AAAI International Conference on Weblogs and Social Media (2008), .
Completely agree, I'd like to add citation of policy as a code point for a future study. Staeiou 17:17, 13 June 2011 (UTC)Reply[reply]
I suspect we will also find that some people who do this are new to that particular wiki but not new to wikimedia - with Single User Login that should be easy to check. I've even come across an editor whose experience had been gained on Wikia. Policies do vary by language and project, but such is the influence of Wikipdia that an awful lot of wikis have big policy overlaps - and where they differ they tend to be conscious of it. I've seen a number of wikis with "How we differ from Wikipeda" pages. WereSpielChequers 13:57, 6 July 2011 (UTC)Reply[reply]


Staeiou & Jtmorgan, could you add some URLs for the Open access, Open source, and Open data nature of this project? You can add them directly to the project template with the open-access-url, open-source-url, and open-data-url params. --EpochFail (talk) 22:50, 9 February 2017 (UTC)Reply[reply]

Just checked and we do have that data in the old wsor shared folder (wsor/stuart/jun8-lifecycles, main data is in data/full_dataset_jun10.xls). The raw dataset contains usernames and userids with our ratings -- are there any privacy issues you can think of EpochFail & Jtmorgan? Any uses of this data that would suffer if we didn't include usernames and userids? Staeiou (talk) 00:54, 15 February 2017 (UTC)Reply[reply]
I don't see any issue with publishing the names/ids, Staeiou. Jtmorgan (talk) 22:59, 15 February 2017 (UTC)Reply[reply]